Colombia - Urban Development Project

Documnt of The World Bank FOR OFFRCIAL USE ONLY Report No. 7210 PROJECT COMPLETION REPORT COLOMBIA URBAN DEVELOPMENT PROJECT (LOAN 1558-CO) April 20, 1988 Infrastructure and Energy Operations Country Department III Latin America and the Caribbean Regional Office |This document has a restricted distributioo and may be used by recipients only In the performance of I a.! Ij .Iv 4 ' ' k * W I *w4 dian  ABBREVIATIONS AND ACRONYMS BP Ranco Popular CAIP Integrated Preschool Care and Nutrition Center CDV Centro de Desarrollo Vecinal (subsequentlv renamed CIP - Centro de Integracion Popular) CFP Corporacion Financiera Popular CT- Centro de Investigaciones de Administracion Publica CULDEPORTES Instituto Colombiano de la Juventud y el Deporte COLCULTURA Instituto Colomblano de Culturs DANCOOP Departamento Administrativo Nacional de Cooperativas DNP Departamento Nacional de Planeacion DRI Integrated Rural Development Project ESAP Escuela Superior de Administracion Publica ICRF Instituto Colombiano de Rienestar Familiar ICCE Instituto Colombiano de Construcciones Vscolares ICT Instituto de Credito Territorial INSFOPAL Instituto Nacional de Pomento Municipal IPC Programa de Integracion de Serviclos y Participacion Comunitaria MAC Basic Module of Health Care MEN Ministerio de Educacion Nacional MINSALUD MiniRterio de Salud Publica PAN National Nutrition Program SENA Servicio Nacional de Aprendizale SIP Secretaria de Tntegracion Popular SuperCoop Superintendencia Nacional de Cooperativas  MRl OmcICAL UK ONLY THE WONtO SANK Wash.ntont. .C.2O l3 U.S.A. Ol6e. d Dwtnrwa CWeof 0msw ieaw#u,A April 20, 1988 MEMORAIMDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Colombia Urban Development Project (Loan 1558-CO) Attached, for information, is a copy of a report entitled "Project Completion Report on Colombia Urban Development Project (Loan 1558-CO)" prepared by the Latin America and the Caribbean Regional Office. Further evaluation of this project by the Operations Evaluation Department has not been made. Attachment | This document has a restricted disribution and may be used by reipiets only in the psrforma_e of their officil duties Its contents .y not otherwis be disclosed witbout World Bank muthoutio.  FI OFfICIAL U8 ONLY PROJECT CCOK.TION REPORT URBN DU VELOPHENT PROJECT LOAN 2558-CO TABLE OF CONTENTS Paxe No. Prefac-*** I................................ B-sic Dat- Shet ................................. Highlights .......................................... .......... ... v II. PROJECT IDENTIFICATION, PREPARATION AID APPRAISAL .......... 2 Preparation, Appraisal, and Negtoiations................ 2 Principal Objectiv ...................................... 3 Project D4scription ...................... 4 Conditions *nd Covenants ...................... . 4 II. IM PLENTATION.... . . ..*** ............ 00....... 4 Effectiveness and Start-Up ....... 4 Revisions ..................... ................................. 5 Implementation Schedule .......................... *....... 6 7 Performance of Contractors end COSUtnt...................... 8 Reporting ....................................................... Project Achievements .............................. 8 IV. FINANCIAL PERFORANCE ............. ......................... 11 Project Cost ............... .... ................... 11 Sources of Finance ...................................... 12 13 Co0t Recov ry ..................... 13 V. ECONOMIC REEVALUATION................ .............. ....... 14 14 15 VI. INSTITUTIONAL PERFORI4ANCE .........................0....... 15 VII. BANK PERFORMANLE..................... 16 VIII. CONCLUSIONS AND LESSONS LEARNED ............................ 17 Overall Performance .......... 17 Lessons Learned......................................... 17 This document ha a rstiedditibudon nd may be usd by recipientsoy In te pfomance of their offc duti its contents way not otewise be disksed whout Word Dank authorIzton.  Table of Contents (cont'd.) PMte No. TABLES 1. Implementation Schedules Appraisal vs. Actu-l.............. 19 2. Physical Achievement ....................................... 20 3. Total Project Cot..............................21 4. Sources of Finance.......................................... 23 5. Schedule of Disbursements ................................... 24 6. Disbursement Allocations ............... .....0............ 25 7. Anticipated and Actual Price Increases During the Project Implementation Period........................ 27 8. Cost Recovery - Shelter Components.......................... 28 9. Cost Recovery - Water Supply end Sewerage Component ......... 29 10. Social Achievements ......................................... 30 11. ReestSmation of the IERR for Sites and Services Investments: The Case ofPerera........................ 32 12. Reestimation of the IERR for Water Supply Distribution Investments: The Case of Pereira.**........ 33 13. Implementation Responsibilities ............................. 34 ANNEXES 1. Project Components as Appraised, Revised and Implemented.... 35 2. Borrower's Compliance vith Key Loan Covenants..............e . 38 FIGURES 1. Administrative Organization of the IPC Pror. . . 39 2. SIP's Organizational Chart.................................. 40  ,(i) COLOSBIA - LOAN 1558-CO URBAN DEVELOPMDNT PROJECT PROJECT COMPLETION REPORT PREFACE This is the Project Completion Report (PCR) for the Urban Development Project, which was partially financed by Loan 1558-CO, in the amount of US$24.8 million, of which US$18.0 million was used in this project and US$6.8 mlIlion was reallocated to the Popayan Reconstruction Project. The Borrower was the Government of Columbia, which transferred the funds, through the national treasury, to the following project executive agenciess DNP, MEN, MINSALUD, ICT, ICCE, ICBF, CPP, SENA, INSFOPAL and SIP. The PCR was prepared by the former Urban Project Division of the Latin American and the Caribbean Regional Office, based on information in the Bank files, a Final Project Report prepared by the Borrower, discussions with staff of the Bank and the Borrower and visits to the field. In accordance with the revised procedures for project performance audit reporting, this PCR was read by the Operations Evaluation Department (OED), but the project was not audited by OED staff. Following normal procedures, tED sent copies of the draft report to the Government and the Executing Agency for cov_ents; however, none were received.  (ii) PROJECT COWPIZTI(W REPORT COIZIA URBAN DEVELOPNENT PROJECT (LOAN i55s 0) BASIC DATA SHEET Key Project Data Apdrai*al Actual or Itewm Estimate Current Estimate Total Project Cost (USS million) 62.0 65,9 Underrun or Overrun (2) a/ 6.3 overrvu Loan Amount (USS uillionY 24.8 24.R Disburoed 18.o Reallocated h/ 6.R Cancelled - Date Physical Components Complete 17/31/1981 12131/1983 Proportion Completed hy above date (2) c/ 30 95 Tine Overrun (2) 55 Financial Performance Moderate Problems I3ntitutional Performance Moderate Problems Economic Rate of Return (2) 10 10 Shelter Components Water Supply Component 10 10 a/ In relation to adjusted project cost of US$63.9 million cost overrun was 32. T/ Loan balance allocated to Popavan Reconstruction Project - Loan 2379-CO. c/ 952 based on adjusted physical targets and appraisal targets due to increase In water supply component under adjusted scope. STAFF INPUT (Staff Weeks/FY) 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 Total Identification- Preparation 17.0 38.0 55.0 Appraisal/ Negotiations 61.1 35.2 96.4 Supervision .7 40.2 56.5 46.9 31.6 1%l.n 13.7 4.R 209.4 Total 17.0 99.1 35.4 40.2 56.5 46.9 31.6 15.0 13.7 4.8 360.7 CUMULATIVE ESTIMATFD AND ACTUAL DISBURSE1ENTS (ITS$ million) FY/79 FY/80 FY/R1 PP/R2 PY/83 fY/fl4 FY/R5 (i) Estimated 7.1 13.5 19.7 24.R (it) Actual 0.7 1.9 5.1 10.6 14.6 18.0 X of (II) to (i) %.2 9.6 20.6 42.7 58.9 72.6 a/ a/ Excludes US$6.8 million allocA,ed to Popayan Reconstruction Project - Loan 2379-CO.  (i11) Other Project Data Item Original Plan Act-ual First mutioned In Files/Tiuetable 04/08/7R Government Application 09/03/75 Appra sal 07176 a/ 02/77 Negotiations 11/29/77 02/28/78 Board Approval 01/17/78 h/ 04/22/78 Loan Agreement Date 07/21/78 Effectiveness Date 12/01/78 Closing Date c/ 06/30/R2 12/31/84 Borrover Government of Government of Colombia Colombla Executing Agencies DNP, MEN, DNP, MEN, MINSALUD, NINSALUD, ICT, ICCE, ICBF ICT, ICCE, ICBF, CFP, SENA, CFP, SENA, INSFOPAL, SIP INSFOPAL, SIP Fiscal Year of Borrower Jen.1 - Dec. 31 Follow-on Project Cartagena nrban Development(1694-CO) Popayen Region Earthquake Reconstruction Project (2379-CO) a/ Revised to 01/77 in First Project Brief dated 10/26/76. b/ Originally scheduled for FY77. After postappraisal mission, Board date was revised and set for 01/17/78. c/ Closing date extended to allow use of funds transferred to Popayan Reconstruction component. However the above date reflects the closing of disbursements for the urban development component, oblect of this report.  (IV) Mission Data Staff Date of Performance Types of Mo/Year No. Weeks No Persons Weeks Report Rating Problem Identiflcation 04/75 1 2 2 04/29/75 Preparation 09/75(*) 1 3 3 10/10/75 02/76(*) 2.5 2 5 03/23/76 05/76 2 4 7 06/11/76 07/76 3 5 11 09/16/76 12/16(*) 2 2 4 01/04/76 Appraisal 02/77 3 7 21 03/22/77 05/77 2 5 7 06/21/77 09/77(*) 1 1 1 09/28/77 Supervision 07/78 3 4 11 08/18/78 1 I 11/78 2 2 4 12/12/78 2 M 01/79 2 3 5 n2/22/79 2 M 05/79 3 1 3 06/29/79 2 M,T 08/79 2 5 10 09/20/79 2 M 10/79 1 1 1 11/21/79 - - 12/79 2 4 7 02/12/80 2 M,F 06/80 2 4 7 06/25/80 3 F,M,P 10/80 2 4 6 11/03/80 3 F,M.T,P 04/81 2 4 7 04/27/81 3 M,T 06/81 1.3 3 4 07/27/81 - - 01/82 1.3 2 3 02/01/82 2 M,O 04/82 1 2 2 04/30/82 2 _ 07/82 1 1 1 08/25/82 2 P,F 12/82 1.6 2 3 12/17/82 2 F 09/83 1 1 1 09/30/83 2 P 05/84 1 1 1 06/15/84 1 F 05/85(**) @ 1 1 06/11/85 3 M,O TOTAL 27 missions 138 * Combined with preparation of Loan 1694-CO ** Combined with supervision of Loan 2379-CO  (v) PROJECT COMPLETION REPORT COLOMBIA URBAN DEVELOPMENT PROJECT (LOAN 1558-CO) HIGHILIGHITS The project has attained practically all its social objectives, having Improved the levels of education, health, nutrition, and marketable skills of family members belonging to the lower half of the urban income distribution in a scale that corresponds approximately to one fourth of the population of the 23 project cities. It reached or surpassed the physical targets for all project components. The project was less successful in achieving Its financial goals and long term Institutional objectives, and faces uncertainties with respect to its replicability and sustainability over time. The Institutional coordination difficulties the project faced throughout the Implementation period could have been more manageable if the project had been designed to include fewer clties and fewer project compo- nents. This would have also possibly allowed a better integration of local municipal governments into project design and the eventual transfer of the responsibility for the operation of the project financed facilities (i.e. Community Facilities and Servlces Centers - CDVs) which was not envisaged and are therefore, still being administered by the Cen.tral Government. Moreover, sustained community participation in most project zones appears to depend on central governments funds suggesting a pattern far removed from the original idea of community managed programs. Also, a more thorough assessment of the management capacities of the executing agencies such as INSFOPAL and ICT at the time of appraisal could perhaps have identified better their institutional weaknesses and through corrective actions under the project could have strengthened thelr respective capaci- ties and performance. The project was envisaged as a first phase of implementation of a national program for the integrated provision of services and community participation (Programa de Integracion de Servicios y Participacion Comuni- taria - IPC), and was to provide the Borrower with the means of refining that program and ensuring its sustained cost effectiveness as coverage expanded. This objective has not been attained because the project falled to develop: adequate management information system and Impact evaluation measures that would give a solid basis for future program choicee; adequate cost recovery in shelter and water supply components, Implying a much higher level of national subsidization; and, satisfactory arrangements for the administration and operation of the Community Facilities and Services Centers. The less than adequate financial performance of the project raises questions beyond the relative compliance with Loan covenants. The history of the project suggests that in order to carry out multi-sectoral programs at the local level, the habitual budgetting and programming programs of the country had to be disturbed and changed.  PROJECT CO(PLETION REPORT COLOMBIA URBAN DEVEOPMENT PFRECT (LOAN 1558-CO I T INTRODUCTION 1.01 Loan 1558-CO financed an integrated urban development project which provided, in a concerted and focussed manner, social services, credit, and infrastructure to a large group of low income households In 23 medium size cities. The project also provided technical assistance and training to Increase employment and productivity among the target popula- tion. The Bank loan of US$24.8 million was made to the Government of Colombia. Project funds were transferred, tbrough the national treasury, to nine project executing atencies. Four agencies, namely Servicio Nacio- nal de Aprendizaje (SENA), Instituto Colombiano de Bienestar Familiar (ICBF), Instituto de Credito Territorial (ICT) and the Banco Popular, financed a total of 45Z (US$29.4 million) of the project costs. The share of the National Government was 282 (US$18.6 m'llion) and the disbursed amount from the Bank Loan (US$18.0 million) financed 272 of total project costs of US$65.9 million. US$6.8 million of the Bank Loan was reallocated to the Popayan Reconstruction Project (Loan 2379-CO) in August 1984. 1.02 The justification for the project rested on urbanization and income disttribution trends observed in Colombia In the mid 1970s. This urbanization pattern was already evident in 1973 when half of the 12.6 million urban dwellers llved In cities other than the four major metropoli- tan areas of Bogota, Medellin, Call, and Rarranquilles High rates of popu- lation growth In intermediate cities reflected rapid rural to urban digra- tion and created deficits of urban infrastructure, housing, and social services that were particularly severe In low-income neighborhoods. On the other hand, although Colombia experienced impressive economic growth during the 1964-1974 period, when GDP grew by an average of 6.52 per annum in real terms, by the mid-1970s the pace of growth had slowed and unemployment increased, especially In urban areas. These trends heightened governmental awareness and prompted the administration which took office in 1974 to launch programs to improve the welfare of the poorest families. One such program was the national program for Integrated provision of services and community participation, the Programa de Integracion de Servicios y Participacion Comunitaria (IPC). A new agency was created In the Office of the Presidency, the Secretaria de Integracion Popular (SIP) to coordinate the program. Both the IPC and SIP were ratified by Presidential Decree in January 1977. 1.03 This report is based on information in the proiect file, includ- ing a completion report prepared by the Borrower, and on additional data, Interviews, and field visits.  -2- II. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL Preparotion, Appraisal and Negotiations 2.01 The project was prepared by the National Planning Department (DNP) in consultation with the sectoral agencies and the participating communities. Bank assistance in this process began in 1975 and contributc- to broadening the scope of the program and phasing its implementation.l/ The selection of the project cities was based on a study done by DNP Tn which 58 cities were ranked as to the incidence of urban poverty, popula- tion jrowth rate, and public service needs. The 23 cities eventually Included In the project exhibited high rates in all these three dimen- sions. The project was appraised on the basis of prefeasihility studies conducted in 6 project zones, on technical standards and prototype models of facilities furnished by each sectoral agency, and on preliminary studies of the remaining 17 cities2/. At the time of appraisal, SIP was newly created and had but miniiail participation in the project preparation process. The appraisal mission focussed on the evaluation of the program model, and several project preparation tasks were left to the implementa- tion phase. This in part explains the difficulties encountered during the start-up phase of the proiect. Two postappraisal misslons, in May and September of 1977, were necessary to complete preparation of the water supply component, finish the implementation plans for the first 6 zones, clarify a number of inetitutional issues including the choice of the project coordinating agency.3/ 1/ At first project design was limited to the provision of community services in the target neighborhoods through community development centers (CDVs). With regard to programming, the national program was initially a 3-year investment program covering 44 neighborhoods in several medium sized cities and del,srtmental capitals. After reviewing its financial implications, DNP decided that the program had to be phased into implementation periods of 3 to 4 years, each serving a smaller number of neighborhoods. The first phase corresponds to the present project in 23 cities; the second phase to the Cartagena Southeast Zone Rehabilitation Project (Loan 1694-CO) and the third phase would extend the program to the remaining cities. / The zones, object of prefeasibility studies were Armenia, Cartago, Cucuta I, Pereira I, Girardot and Villavicencio. 3/ The first proposal from the Borrower was to divide project coordination between DNP and SIP, DNP being responsible for the physical and the environmental components and SIP for the rest. After the Bank expressed its preference to relate to a single coordinating agency, the Borrower selected SIP to perform this function.  2.02 There were deliberations within the Bank regarding the best was to reconcile the urban project with other operations that overlapped In time and space in the water supplv and small scale Industry sectors.4/ Concerning the employment and productivity componenca, deliberations withTn the Bank centered on the issue of the volume of Industrial activity in some project cities that was too swall to justify vocational training and credit operations, suggesting that it would be unwise to insist on minimum lending and training targets per city.5/ This highlighted the need for ensutring that all executing agencies have full responsibility for implementation of their respective components and adjustments thereof, as well as for system- atic project monitoring and evaluation. Other lnstituttional arrangements disciussed, pertained to the ootentiai overburden of sectoral agencies by the simultaneous execution of varlous externally financed operations, which was considered a risk. The Inexperience of SIP and the location of that agency at the Presidency was also considered a risk. 2.03 Negotiations scheduled for November of 1977, were postponed at the renuest of the Borrower until certain issues were resolved to the satisfaction of the Ministry of Finance. These Issues were: (1) budgetary mechanisms for funding project expenditures and procedures for submitting loan disbursement applications to the Bank; (it) agreement on draft project contracts between sIP and the participating agencies; (11i) Improvements of SIP's managerial capacity through appointment of a technical director; (iv) clarification of required contents of area Implementation plans. Negotla- tions took place in February, 1478 In Washington, n.C, and Board presenta- tion on April 22, 1978. Principal Objectives 2.04 The maior objective of the prolect was to improve the living standards and productive capacity of about 550,00n Colombians living In relative poverty in 23 citles throughout the country.6/ Strengthening the managerial capacity of SIP and of participating agencies to plan and execute the program was a second lmportant Project objective. Although the project does not include a technical assistance component with this aim, operational expenses incurred in biripg and training project staff for key executing agencies were considered integral part of the project cost. The third important objective was that the project would enable the Gove-nment, through systematic monitoring and evaluation, to refine the IPC program and assure Its sustained cotit effectiveness as coverage expanded, both during the life of the project and thereafter. 4/ The Second Small Scale Industry Project (loan 1451-CO) overlanped with the urban project In Pereira, Cucuta, Ibague, Pasto, Bicaramanga, Armenia and Manizales. The Water Supply loans P60-CO and 1072-CO covered 15 cities of which only two were not also Part of the urban project. The overlapping cities were: Cucuta, Villavicencio, Armenia, Manizales, Neiva, Pereira, Tulua, Santa Marts, Ibague, Pasto, and Valledupar. 5/ See commeats on he Yellow Cover Appraisal Report made by J.T. Rentley, IBR1, 04 dated 08/12,1977. 6/ See, IBRD, SAR-Colombia, Urban Development Project, April 10,. 1q78, p.l0.  -4- Pro,ect Description7/ 2.05 Loan 1558-CO included investments in 27 zones located in 23 cities covering the following subprojects: (1) improved social services, including the construction of and equipment for comunity development centers, primary schools, health centers, and child and family welfare centers (CAIPs), as well as training of teachers, community health and child welfare social workers; (it) increased employment and productivity including managerial assistance for promotion and organization of private and cooperative enterprises, vocational training programs, and credit to small scale Industries; (iii) physical and environmental Improvements, including area upgrading through the regularization of land tenure, credits for neighborhood and home Improvement loans, sites and services develop- ment, and basic infrastructure services such as water and sewerage; and (iv) project management and evaluation, lncluding strengthening the imple- menting agencies by financing appropriate studies and surveys; and strengthening the coordinating agency by funding initial operating expenses and training programs so that It would he hetter able to monitor, evaluate and refine the project. Conditions and Covenants8/ 2.06 The covenants negotiated reflected the importance of: the timely and sufficient provision of counterpart funds by the Borrower and the project entities; the effective recovery of water supply, housing and small scale industry credit costs; and, the sustained Institutional support for the project. Specific loan covenants also established that general guid- ance to the program was to be provided by a National Council formed by representatives of DNP, SIP, and the participating agencies and that monitoring and evaluation would he by SIP. 2.07 The conditions of effectiveness were signing of the agreement between Banco Popular and CFP; and signing of the agreements between the Government and the other project entities and between SIP and the other participating ministries. In addition, a condition of disbursement in each project zone was completion of the implementation plan for the respective zone satisfactory tc the Bank. III. IMPLEMRNTATION Effectiveness and Start-Up 3.01 The project was declared effective December 12, 1978, with about two months delay. Conditions of disbursement were met only after signifi- cant delays because of SIP's inability to complete Area Implementation Plans (AIPs) satisfactory to the Bank. The first 16 AIPs were approved in 1979, about one year after the loan became effective. One plan was approv- ed in 1981, and the remaining 6 plans were approved only in 1982. 7/ Detailed project description is given in Annex I. 8/ Major Loan Covenants are listed In Annex II.  3.02 After effectiveness, all project execucing agencies with the exception of SENA and -- to a lesser extent, TCT in the CDV component, were slow to begln project execution. The reasons for the initial delays were Interrelated, namely: the project's extensive geographical coverage; the large number of proiect executlng agencies; inadenuate budgetary support; inexperienced staff and hlgh staff turnover at SIP; political constraints derived, ln part, from SIP's location In the Office of the Presidency; and lack of understanding about the contents of the project at the local level.9/ Detailed engineering for water supply iavestments did not start until Tate 1979. The contracts for the first CDVs were negotiated by ICT on the hasis of preliminary designs, not adjusted to the sites, causing construction delays and serious construction defects that required exten- sive repalrs and, in Girardot, the total reconstruction of the facility. The loan from Banco Popular to CFP was signed but not disbursed until mid-1980. Reliable financial and physical progress reports were lacking because SIP had not been able to establish and maintain systematic manage- ment Information systems prior to mid-1980. As a result, by June, 1980, almost two years after Loan effectiveness, an estliated 12% of total project investments had been completed, compared to 442 projected at appraisal. Disbursements of the Rank loan were even less, totalling US$0.75 million, or 52 of the appraisal estimate. Implementation delays and the several dilfficulties encountered during this period, prompted discussions between the Rorrower and the BAnk regarding the eventual revi- sion of the scope, content, and schedule of execution of the project. Deliberations on these issues started during the second half of 1980, lasted through 19R1, and were concluded in early I9R2. Revisions 3.03 Many of the components under the project were sirnificantly revised in May 1982 (Annex I).10/ Delays in finalizing project revision. were caused by difficulties in prrogramming each component in the absence of adequate data, the need to complete detailed engineerlng for the water supply component, and the time required to revise all project agreements between SIP and the executing agencies and between INSFOPAL and the Munici- pal Empresas.'1/ The revisions made did not substantially alter the scope of the proiect given that the number of components and cities remained unchanged and only four subsidiary project zones were eliminated. However, 9/ During the start-up period of project Implementation, from mid-1978 to November, 1Q80, SIP's director changed five times each time occasioning also the change of key central staff. 10/ As per Amendment to the Loan Agreement, formelized by letter dated May 13, 1982. 11/ The revision of physical and financlal targets for each component were based on: (i) more realistic demand figures by city provided by project executing agencies; (ii) improved cost estimates and investment schedules whict. resulted from the completion of detailed engineering for the water supply component; (iii) the demonstrated implementation capacity of project executing agencies and limitations thereof; and (iv) cost increases affecting contracts for civil works, equlpient supply, and professional staff salaries, whlch resulted from domestic inflation In excess of appraisal projections.  -- the revlslons altered the physical and financial targets of almost all components and provided for additional strengthening of SIP and for provi- sion of funds by the Natilonal Oovernment and the project entities.1Z/ The substantlal reduction In the houslng couponents wea prompted by the labil- Ity of ICT to liplement the components efficlently In sose project cities, that In turn was associated with Inadequate funding and stafflng of ICT local offices. The Increase in water supply Investments reflected the higher politlcal priorlty attached to urban infrastructure at the tle, as well as progress made by INSFOPAL In procurement of supplies and engineer- ing. Only one change proposed by the Borrower was not approved by the Bank, and that was the introduction in the project of a water treatment plant in Sahanalarga, disallowed because Its financial feasibllity was not demonstrated to the satisfaction of the Bank. The Borrower and the Bank were in agreement with respect to all other changes. 3.04 The above revisions resulted in addltional social benefits accru- Ing from higher primary education and child care and nutrition service levels as well as from expanded training programs carried out by SENA. The prolected additional benefits resulting from the increment In water supply investments, were only fully realized two years after Project completion due to execution delays which affected the installation of domestic water connections in 10 of the 21 cities where water supply investments were made. In retrospect, project adjustments could have been more effective by also addressing the project's excessively large and dlspersed geographical scope; the long-term arrangements for the operation and maintenance of CDV facilities; and by Insisting on the application of cost effectiveness and financial viability criteria to key project programming decisions. 3.05 A second revision of the project occurred In 19i4, when savings of US$6.8 million of the Loan was reallocated to the Popavan Reconstruction Project (Loan 2379-CO).13/ This reallocation was requested in March 31, 1984, in view of saviigsa resulting mainly from the devaluation of the Colombian peso in relation to the dollar. Implementation Schedule 3.06 The actual implementation schedule lagged far behind appraisal projections during the first two vears of project execution. (Table 1). Implementation also differed from appraisal in terms of the sequentlal order of completion per city. From Loan effectiveness, the prolect took five years to reach full completion. Disbursements were closed as of December 31, 1984. However, the Loan remained open until 1996 to allow disbursements under the Popayan Reconstruction Project. 3.07 Although all project executing agencies experienced some diffi- culty during inplementatlon of their respective components, most of them were able to compensate for the initial delays and complete their part of the project within the revised schedule. The exceptions were: (i) INSFOPAL, which completed the water supply component two years after the revised project completion date; (ii) ICT, which completed construction of 12/ Budgeting and transfer procedures were modified to ensure more adequate and timely funding of project expenditures. 13/ See, IRRD Loan 155R-CO, Amending Loan Agreement, dated August 2, 1984.  -7- the last two CDVs in San Andres and Riohacha only in 1984; and (iII) the Ministry of Realth, which has not yet been able to train the number of paramedical staff projected at appralsal. Lack of adequate and timely fundlng by the National Government contributed significantly to the delays. However, delays were also due to administrative Inefficiency and tardiness In the preparation of detailed engineering and contracts with Munlelpal Water Companies, on the part of INSFOPAL. Lack of appropriate staff and management continuity, affected the performance of both IC!T and INSFOPAL during the project life. With regard to SIP, inedequate and frequently changing higher management staff and the lack of experience in coordinating multi-sectoral projects, adversely affected its implementation monitoring ability during the first two years, suggesting that the modest achievements of the project during that period were entirely due to the implementation capacity of the project executing agencies. Coordination improved substantially with strengthened management and staffing of SIP that occurred as of 1981. Rowever, the two vear delay in implementation already accumulated was only in part compensated by the subsenuent acceler- ation of project investments. Procurement 3.08 The Borrower procured all Bank-financed goods and services in accordance with Bank guidelines. In line with appraisal expectations, most contracts for civil works, furniture, supplies, and consulting services were awarded to local firms. Most equipment was purchased from foreign suppliers. INSFOPAL was able to save costs by procuring in bulk, at an early phase of project implementation, pipes, fittings, and water meters necessary for a suhstantial portion of the water supply component. Seven main supply contracts signed by INSFOPAL were awarded on the basis of International competitive hidding between 1979 and 1981. Most civil works contracts were awarded on the basis of local competitive bidding with the exception of small works, especially for school renovation, which were negotiated directly wlth local firms or, In the case of Municipal Empresas, carried out by force account. Procurement procedures used in the water supply component were cumbersome, requiring reviews by INSPOPAL, Municipal Water Companies, and by the Bank. In retrospect, procurement decisions made by the Borrower and approved by the Bank with respect to ICT civil works contracts for CDVs were unwise.14/ These contracts were awarded, with Bank approval, without prior compThetion of detailed engineering and soil analysis, through direct negotiations by ICT with local firms. As a result, the first 6 CDVs built suffered serious defects. The component in general incurred in substantial cost overruns, delayed execution, and faul- ty construction requiring extensive repairs. Correct procedures involving local public bidding and standard bidding documents were iutroduced later in connection with the construction of the last CDVs In 1983 and 1984. 14/ The Bank accepted ICT's judgement that reliable contractors would be unlikely to bid competitively on construction contracts of the CDV facilities in project areas. Therefore, contracts amounting to US$8.2 million, were carried out by ICT through the direct administration of small local contractors chosen In accordance with ICT's tendering procedures. See, IBRD, SAR, op.cit. Para. 5.17.  Performance of Contractors and Consultants 3.09 With the exception of the contractual problems described in Para. 3.08 above, and the delays experlenced in finalizing contracts for civil works and engineering for the water supply component, the performance of consultants and contractors in the project was satisfactory. Problems encountered in the evaluation consulting contract cannot be attributed to the consultant and are discussed In Para. 3.15 bPelow. Reportlng 3.10 Timely, comprehensive and reliable reports were not received during the first two years of project implementation. Guidance from Bank supervision missions as to the selection of key indicators for project monitortng -s 4off4ered to SIP only in 1980, two years after Loan effective- ness. Accounting procedures were substantially improved after 1Q80 but remained less than adequate In the case of ICT, as indicated by the fact that in December 1981, 65Z of ICT investments in sites and services and 35% of its Investments in home improvement loans could not properly be account- ed for. These accounting deficiencies also demonstrated inefficient finan- cial management on the part of SIP during part of the project life. Audit- ed accounts for all project entities were submitted to the Bank only after considerable delay. Project Achievements 3.11 The project achievements reached or surpassed the revised targets set for almost all components. Table 2 describes these achieve- ments in detail and compares actual results with both the appraisal and the revised program.15/ Only two cities, San Andres and Riohacha, failed to attain full compTetion of the social services subproject. In all others, community development centers, educational, health and child care and nutrition facilities were built and equipped to appraisal standards or higher, and were fully utilized by project completion date. With regard to service levels attained, a deficiency is apparent In the area of health, because only 25% of the paramedical staff programmed was in fact trained by the Ministry of Health and Incorporated Into the project. These trained staff, are working at the health centers and not In community outreach programs as originally planned. Other variations in actual service provi- sion adhere to the original obJectives of eacb subcomponent and are described in detail In Annex I. 3.12 SIP actively promoted community participation at the local level, but the results of these efforts were uneven. The community support group created In each CDV failed to effect community management of the program. Community groups that were already organized before the project began, such as the neighborhood actlon committees (Juntas de Accion Comunal), were often bypassed In favor of new groups directly organized by SIP. Among the groups organized by SIP, those that received funds for specific activities tended to function best, as for example the sports committees. In cities where the community has participated intensively, as 15/ City specific achievements are given in the Borrower's Project Completion Report, available in the project file.  -9- for Instance in CLicuta and In Pereira, some of these specialized groups were able to mobilize addltional financial resources for the program.1 / In the most successful cities, comunity participation led to improvements In the relationships between mnmicipal authorities and SIP's local staff. 3.13 The less than expected disbursements of small scale enterprise loans reflects, in part, lower actual demand for such loans, given the small size of the industrial sector in some of the project cities. On the other hand, CFP had difficulty in lending in cities where it did not have an office. Furthermore, because CFP cannot lend for service and commercial enterprises and requires collateral for all its loans, several small scale entirprises in the project areas did not qualify for the credit program. Partial Improvement occurred when a National Loan Guarantee Fund was made available for use as collateral for CFP loans. The performance of SENA's vocational and managerial courses was impressive on several counts: the early start of this activity and its consistent progress during project limplemwntation; the ability to adjust course programs to local needs; and the results achieved In terms of number of graduates was 30% higher than the revised targets and more than twice the appraisal proJections. However, because SENA did not undertake tracer studies of graduates, It is not possible to ascertain the exact employment impact of this program. Never- theless, adjustments made in course content during implementation Indicate that SENA qualitatively monitored the local labor market and the relevance of its courses and labor demand. Finally, the experimental cooperative program was the least successful as it did not generate formation of any self-sufficient cooperative. 3.14 She physical and environmental improvements subproject achieved the revised targets for housing. On the other hand, the land tenure regu- larization component was not implemented, since the regularization of land tenure is the responsibility of municipal governments and not of ICT. The reduction of targets for land tenure regularlzation made in 1982, did not resolve the problem because ICT was unable to hire the legal and survey staff required to effectively assist municipal governments in this respect. As a result, the property titles issued to project beneficiaries were obtained through their own effort and expense. In the water supply component, the revised targets for investments in water distribution systems, special works, and household metered connections in 21 cities were attained in 11 cities by December 1983 and in the remaining 10 cities by 1985. This less than satisfactory performance by INSFOPAL and Municipal Empresas affected more severely the subprojects carried out in Santa Marta, Riohacha and San Andres which began during the last year of project imple- mentation. Sewerage investments, although mentioned in the appraisal, were not in fact programmed. Nevertheless, extensions to sewerage collection 16/ Typically, SIP would organize specialized committees in the areas of culture and sports; household improvements; production and employment; and infrastructure. See Borrower's PCR, op.cit., in the project file.  -10- networks benefitting project zones were made In two cities after water investments were completed.17/ 3.15 In contrast with the relatlvely high physical achievements of the project, the performance of the evaluation component has been disap- pointing. The evaluation study Was subcontracted with Instituto SER, a private research organization based in Bogota of excellent standing. The actual contract and Its Initial funding however, were delayed and the first three baseline surveys were carried out in project areas only in late 1980. Flnancial constraints In SIP led to the cancellation of this consulting contract ohortly after the completion of the baseline surveys. although the data collected In these surveys is useful to demonstrate that the project has Indeed reached the poorest residents in each city and that economic conditions in each city played an Important role In the success of the employment and productivity components, the cancellation of the second and third sets of surveys planned, meant that measures of change were never undertaken. As a result, the economic and social impacts of the project cannot he systematically analyzed.18/ 17/ See, INSPOPAL, Programia Zonas Marginales, Convenio 1558-C0/BIRF, Infor- me de Avance, May 1986, In the project file. The observed delays ln the execution of this component might in part he attributable to over- burden of INSFOPAL and the MunIcipal Empresas by the simultaneous Implementation of other externally financed projects. For example, the Third Medium Size Cities Water Supply and Sewerage Project (Loan 1726-CO) overlaps with the urban project in Riohacha, Cartago, Sabanalarga, G.irardot, Ibamue, Neiva, Pasto and Sincelejo; arid IDB loau for water supply Implemented during the same period, overlaps with the urban project in Barrancabermela. 18/ The final evaluation report submitted by Instituto m is in the project file. Also available is a paper by Dani Kaufmann, using data collected in the baseline surveys, entitled, WA Study of Marginal Areas within the Context of an Integrated nTrban Project: The Case of Three Cities Included In the Colombia I Urban Project - IPC", Draft, IBRD, March 1981.  -11- IV. FINANCIAL PERFORMANCE Pro4ect Cost 4.01 Project costs are shown bv component In Table 3*19/ Total cost exceeded the revised and the appraisal estimates by 6.3X ai; 6.42 respec- tively. The original cost structure has changed in that both the revised and the actual figures show a greater emphasis on social services. Higher project management costs reflect the decision to strengthen project coordi- nation through SIP. The relevant sumary figures are given helow. ACTUAL AND ESTIMATED SHARE OF SUBPROJECTS IN TOTAL PROJECT COST (in current prices) Appraisal Revised Estimate Estimate Actual SubproJects US$M 2 USSM x US$M X Social Services 14.2 23 20.3 32 22.0 34 Employment and Productivity 7.8 12 8.0 12 8.3 12 Physical and Environmental Improvements 35.9 58 29.8 47 29.9 45 Project Management and Evaluation 4.1 7 5.8 9 5.7 9 Total 62.0 100 63,9 100 65,9 100 4.02 For the components that experienced greater cost variances, an analysis of unit costs indicates that ICT incurred signiflcant cost over- runs in CDV construction and so did ICBF with respect to facilities for child care services. In the case of ICT, these cost overruns were due, for the most part, to management Inefficiencies; while In the case of ICMF, the main reason was that standards used were consistently higher than the original plan. The pertinent figures are shown helow. 19/ Costs per city are detailed in the Borrower's Project Completion Report, oo.cit., in the project file.  -12- ACTUAL AN~D ESTIMATED UNIT COSTS FOR SELECTECD C0ONPONENTS (in current prices) Aopraisal Revised Estimate Estimate Actual Selected Components USA X US$ S Variance US$ X Variance Coneunity Development Centers (per CDV) 314,815 100 391,304 24 471,429 S0 Child and Fatly Welfare (per child) 136 100 318 134 355 161 Vocational Training (per graduate) 134 100 81 (-39) 78 (-42) Small Enterprise Credit (per loan) 2,222 100 2,508 13 3,194 44 Rome Improvement Credit (per loan) 712 100 1,46 110 1,517 113 Sites and Services (per lot) 1,192 100 2,422 103 2,556 114 Sources of Finance 4.03 Table 4 shows a significant reduction in Bank participation in total project financing, from 40X to the actual 27X level. Concomitantly, increases occurred In the financial participation of ICT, SENA, and ICF. The main reason for this large deviation from the original financial plan is due to the delay In processing of loan dlsbursement applications on the part of SIP and the project entities. Since the expendltures on child and family welfare services were not reimbursable, the execution of this compo- nent to higher standards caused ICF to finance a larger share of total proJect cost. 4.04 Not accounted for in total project cost and financing, are addi- tional contributions to the project uade by the Government of Holland, the Government of Canada through CIBA, and by UNICEF. A bilateral aid agree- ment between the Government of Colombia and the Dutch Government In the mid-1970s, provided for SENA's vocational training equipment for workshops In Pereira, Cartago, Cucuta, and Villavicencio. Under the mae agreement, SENA received technical assistance from three Dutch consultants with exper- tise In institutional management, community promotion, and employment generation. This technical assistance progra was completed in March 1979, and probably accounts In part for the successful performance of SENA slnce the start-up phase of the project. Likewise, CIDA provided technical assistance to INSFOPAL In 1981, Indirectly assisting In the management of the project's water supply component. Pinally, a non-quantified amount of assistance was granted by UNICEF to ICBF during the implementatlon of the chlld and family welfare piogram under the project.  -13- Disbursements 4.05 At closing of accounts in December of 1Q84, onlY 11SA1R mlllion of the USS?4.8 million Rank loan was dishursed, and approximately USS6.8 million was allocated for the Popayan component. Domestic Inflation and devaluation of the Colombian peso In relation to the dollar have not been so great as to explain the shortfall which occurred in loan disbursements. 20/ As shown in Table 5, dlsbursements lagged throughout the life of the project. Rather, the main cause of both the lags and the final shortfall In disbursement of Loan 1558-C0, excluding the Popayan component, seems to lie on the complexity of the financial management task assigned to SIP, involving the supervision of a large number of project accounts dispersed amongst nine project executing agencies and 21 Municipal Water Companies. SIP was unprepared and was never able to dedicate sufficient attention to the problem during the life of the project. Although during the last year of project implementation SIP did provide assistance to project exectitlng agencies which led to an acceleration of loan disbursements, the measures taken were too late to compensate for previous lags and Inaccuracies. The financial expertise ultimately assembled in SIP In the early 1980., was used to a greater extea.t to ensure adequate and timely domestic funding for the project, than to monitor the preparation of loan disbursement applica- tions. (Table 6). Cost Recovery 4.06 In general the cost recovery objectives of the project were not met. Among the project executing agencies only CFP was able to reach the level of cost recovery expected under Loan 155R-Cn. The paucity of reliable cost recovery data is also an indicator of inabilitv, on the part of SIP, to vroperly monitor this aspect of the project. Rank supervisIon may also be rated less than adenuate with regard to cost recovery, in spite of the repeatedly expressed concern over ICT's collection performance.21/ 4.07 The cost recovery performance of CFP was satisfactory in so far as the level of compliance by that agency with loan covenants governing lending terms. CFP charged 182 p.a. interest rate for its loans to small scale enterprises, which was sufficient to cover the costs of Its own funds plus operational expenses. Recent data on collections presented by CFP to the project completion mission, show an average cumulative arrears of 11.7% over amounts due in the proiect accounts, considered by CFP in line with the performance of its other similar operations during the some period.22/ 20/ Table 8 in this report. 21/ The only cost recovery covenant systematically monitored by Rank supervision missions was Section 4.06 of the Loan Agreement which regulates the pricing of health services to cover consumable materials costs, notwithstanding the fact that these costs are excluded from project cost. 22/ CFP did not furnish annual data per project city that could be included in this report. However, arrears figures for Loan IR34-CO (Third Small Scale Industry Project) show arrears as 2 of total portfolio varying from 11.4% in 1981 to 17.32 in 19R3. See, LAc-IDFD2, Project Completion Report - Loan 1834-CC, Annex 14, Draft, undated.  -14- 4.08 Cost recovery of home improvement And sites and services loanc hy ICT is significantly lower than appraisal projections throughout the imple- mentation of these two components. Interest was charged at the rate of 122 p.a. or 1.8% below the recommended minimum of l3.RT p.a. In addition, ICT Introduced grace periods for the repayment of its loans, that were not contemplated in Schedule 6 of the Loan Agreement. Year by year collection performance data are not available, but information on a sample of 6,744 accounts In the project zones (Table 9) Indicates that In early 1986 as much as 67Z of all accounts were In arrears and, on these average arrears was 8.6 months. The financlal position of ICT itself has been precarious during project implementation, andt the agency has experienced budgetary constraints that made it diffIcult to staff local offices with the finan- cial personnel necessary to Improve collections. Not much has chanced since then and prospects for improvements in the near future are not good. 4.09 INSFOPAL failed to comply with Schedule 7 of the Loan Agreement, which sets forth the lending terms hetween that agency and the Municipal Water Companies. Collections on INSFOPAL loans are below acceptahle levels In all cities except Pereira. Three Nunicipal Water Companies -- Sabanalarga, Ouibdo, and Turho, were not paying either principal or inter- est on INSPOPAL loans by 12/31/1985. In general, the average arrears was 63% over principal due and 562 over principal and interest due for 21 cities during 19R5 (Table 10). Information regarding the periodicity and the levels of water rate adlustments In the project cities during the life of the project Is limited to Cucuta, where a tariff analysis was made on occasion of project revision when the decision to include or not a water treatment plant in that city's investment program had to be made. V. ECONOMIC REEVALUATION General 5.01 The project was concelved as a process of social change and as such most of its benefits were expected to accrue over the long term and were not quantifiable. A review of the social acblevements up to comple- tion (Table 11) suggests that benefits already accrued to a larger group of urban poor than originally intended. Primary education, chlld care and nutrition service levels were raised far above the original expectations, although less than expected gains were recorded for health services. Benefits from vocational and management training have contrlbuted to the development of marketable skills among a slgnificantly lhrger number of workers than projected, and are helng used in small scale enterprises in the project areas as well as in jobs outside those areas. An estimated 2,100 new lndustrial jobs have been created in conjunction with the small scale enterprise credit Program, or an average of 2.3 new lobs per enter- prise; half of these jobs are concentrated in Ibague, Pasto and Pereira. Benefits derived from housing credits are significant when estimated in terms of imputed rents (Table 12). Water supply investments henefitted a large number of low-income familles through increased comfort, convenience, health, and incremental land values associated with metered water connec- tions. Achievements under this component reached almost 802 of appraisal targets by 1983 and surpassed these targets by 352 two years later. The  -1 S- only expected benefit not realized inder the project is that associated wtth the regularization of land tenure in view of the non-implementation of that component. IERR 5.02 The economic rate of return for shelter Investments, which was recalculated based on the methodology used at appraisal is estimated at 10.92 for Pereira (Table 12). Possibly for the other prolect cities, which tend to be on average less successful than Pereira In project implomenta- tion, the IERR is estimated to be nomewhat lower than 10X for both sites and services and home improvement investments. In future projects which include clearance and resettlement, the analysis of the IERR should take into account costs and benefits of the houses cleared from the squatter areas. For the water supply component (Table 13), the assumptions used at appraisal were verified and the IERR reestimated using the same methodolo- gy, at 10.42 for Pereira. Wowever, financial rates of return, calculated annually for the 1976-1982 period by Municipal Water Companies In 11 project cities are on the whole low. VI. INSTITTTIONAL PERFORMANCE 6.01 The institutional arrangements chosen for the project relied upon the demonstrated Implementation capacity of nine sectoral agencies which were assigned responsibility for the execution of the project components (Table 14 and Pigure 1). Project coordination was the reaponsibility of SIP, a central secretariat created at the Office of the Presidency. It was recognized at appraisal that the uniquely comprehensive design of the project and its extensive geographical coverage, would imply, on the part of the project coordinating agency, an lmportant role in programming the specific responsihilities of the sectoral agencies so that their actions would be coordinated in space and time. In view o' SIP's weakness and inexperience in managing developrvsnt projects, the following measures were included in the project to strengthen SIP's coordination capacity: (i) assistance from DNP and the Ministry of Finance in preparing coordinated annual programs and budgets; (ii) a significant increase in SIP's profes- sional staff; (iii) a review process Including representation of DNP and the participating agencies on the national and local levels ttrouxh the establishment and maintenance of CDV councils; (iv) a monitoring and evaluation component; and (v) further strengthening of project implementa- tion capacity within the various implementing agencies. In addition, It was estimated that in its early stages, the project would require extensive Bank supervision, amounting to about 75 staff-weeks during the first year, and gradually diminishing during the subsequent years of project execution to an average of 49 staff-weeks oer year for the four years of project implementation.23/ 23/ See, IBRD-SAR, Loan 1558-CO Para. 6.16.  -16- 6.02 The institutional analysis undertaken at the ttme of proiect revision,24/ explained how under'taffing, inexperience, and high staff turnover at SIP had seriously hdmpered the implementation of the project. By late 1980, the project had had four Fxecutive Directors in two vear", and the number of SIP staff averaged 10-25 persons compared to 60 persons estimated at appraisal. Two sectoral agencies, INSFOPAT. and ICT, which together accounted for 71% of project investments, were found to have similar staffing problems, particularly in the ICT unit in charge of the shelter components. In order to strengthen staffing weaknesses, project revisions provided for increases in project administration allocations to a level 40% above appraisal estimates. At SIP, emphasis was placed upon the areas of engineering, flnance, loan dishursements, physical planning, management Information systems, and on the intensification of the training program for local CDV staff. Speciflc recommendations for improved hudget- ing and funding procedures were implemented subsequently, and so were changes in procurement procednres for CDV construction. The recommenda- tions for the internal reorganization of SIP, made earlier hy CINAP/ESAP, were finally implemented in November 1980, when a professional Executive Director was appointed and the complete central staff was hired.25/ (See Figure 2). During the following 1981-1985 period, SIP had three Rx-cutive Directors and was able to attract competent professional staff at the central level. The qualitv and number of the staff at the local level was also improved, although not all CDVs were fully staffed by project comple- tion date. 6.03 A gradual transfer of CDV management responsibilitles to depart- mental or municipal governments would be desirable. This wonld mean that the operational costs of the CDVs presently the responsihilitrv of the central government would have to be assumed eventually by local govern- ments. Thus, the sustainahility of CDVs may require contintued subsidies from the national governments, in the medium term. VII. BANK PERFORMANCE 7.01 Bank supervision of the project averaged 10 staff weeks during the 1979-1Q85 period and only in 1980 reached the recormended level of 49 staff-weeks per year. The intensive supervision programmed for the early years of the project was not realized and supervision quality suffered from lack of continuity in the Bank staff involved, as indicated by four shifts in project officers between loan effectiveness and project revisions. As a result, uneven Bank support in the first two years of project implementa- tion, contributed in part to delays in the resolution of project implemen- tation problems. 24/ See, "Organization, Management, and Performance of the Borrower," Annex 9, Supervision Report Loan 1558-CO, April 25, 1991. 25/ SIP's reorganization was a product of a study conducted by the Centro de Investigaciones de Administracion Publica (CINAP) and the Esciela Superior de Administracion Publica (ESAP) in mid-197q which resulted in Resolucion No. 678 of October 23, 1979 that provides for a new organizational structure for the agency. (See Figure 2).  -17- VIII. CONCLUSIONS AND LESSONS LEARNED Overall Performance 8.01 The project has attained practically all Its social objectives, having improved the levels of education, health, nutrltion, and marketable skills of the target population in 23 small and medium size cities. The prolect was less successful in achleving its financial goals and long term Institutional objectives, and faces uncertainties with respect to its replicahility and sustainability over time. Lessons Learned 8.02 (a) Proiect Desien and Institutional nevp et. The objective of strengthening SIP's ability to coordinatMe-he act v ties of sectoral agencies for the efficient delivery of seivizes to the urban poor has basically been achieved but only after seriowo diificulties were overcom- ed. Many of the difficulties encountered by SIE are due to shortcomings in project design. The first design defect relates to the location of SIP, which affected the selection and the stability of its staff stemming from its location in the Presidency and could have been avoided, for example, if SIP was located at DNP. The location of SIP in DNP could have obviated the need for the agency to acquire experience in multi-sectoral coordination, strengthened its planning capacity, and probably also reduced project management costs. In general, institutional coordination difficulties could have been less through the initial selection of fewer cities and fewer project components. Also, a more thorough assessment of the manage- ment capacities of the executing agencies such as INSFOPAL and ICT at the time of appraisal could perhaps have Identified better their institutional weaknesses, and through corrective actions under the project could have strengthened their respective capacities-and performance. On the other hand, the cholce of executing agency for each component was, in most instances appropriate. Only one component - land tenure regularization - was not implemented because of inadequate choice of executing agency. Finally, proJect design did not anticipate the problems of administering local services through a centralized secretariat and therefore, failed to plan for the transfer of the responsibility for the operation of the CDVs from SIP to municipal governments; these weaknesses could perhaps have been considered and corrective measures Included in the project revision made In May 1982. 8.03 (b) Community Participation. Although not an objective in itself, community participation was considered an essential element for the achievement of the project's obiectives and the eventual transfer of the program management to the community was stated repeatedly as a desirable goal. The results achieved in this area are uneven. Sustained community participation in most project zones appears to depend upon grants from SIP to fluance local sports and cultural events, suggesting a pattern far removed from the original idea of community managed programs. On the other hand, as the project completion mission was able to ascertain, local project activities and good CDV management, have attracted the participa- tion of many public and private entities that were not formerly attached to the project. These entities are investing in the project zones or other- wise contributing to further raise the welfare levels of the population.  -18- 8.04 (c) Project Replicability. The proiect corresionds to the first phase of implementation of the national IPC program and was to provide the Borrower with the means of refining that program and ensuring its sustained cost effectiveness as coverage expanded. In general, this objective has not been attained for several reasons. First, the project failed to develop adequate management information systems and impact evaluation measures that would provide a solid basis for future program choices. The lack of such adequate Information has also male It difficult for SIP to adjust the origlnal project model to each elty and, as a result, the project as implemented may not constitute the most cost effective design for some cities and may become less so as these citiea expand and new needs arlse. The second factor 18 the limited cost recovery attained in the shelter and water supply components, implying a much higher level of national subsidization for the IPC program than originally envisaged. And the third factor haspering project replicability Is the unresolved problem of CDV administration. The CDVs are costly to operate and probably overde- signed for the purpose intended. Nore importantly however, they are over- designed in relation to the administrative and financial capacity of most municipal and departmental governments. Thus, while SIP continues to operate the CDVs, It will function less efficientlv and will have more difficulties in expanding the program to other cities and areas. 8.05 (d) Financial Planning and Fiscal Sustadnabllity. The less than adequate financial performance of the project raises questlons beyond the relative compliance with loan covenants. The history of the project suggests that in order to carry-out multi-sectoral programs at the local level, the habitual budgeting and programming procedures of the country had to he disturbed. Thus less than adequate accountabillty and funding were permanent risks during project execution. Because these areas required continuous attention, the planning necessary for long term fiscal sustain- ability was neglected. As a result, the National Government has had to continue to subsidize the operation of some of the project's facilities. Although after loan closing the IPC program in the 23 project cities continues to operate with national funds, further expansion of the program remains uncertain. Not surprisingly, the basic uncertainty lies in the difficulty of adjusting the program to the norms set forth In the Fiscal and Administrative Reform of January 1986.  - a - S - : - : - .- 1 a I - -- . a j-* ||a -L- '111.illlu LI11ll" a~~- ___X _ __X_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __E_ _ _ _ _i _ _ _ _ I i {i .3 ~~~~~~~~~~~~~ . I~~~~~~~~~~l i **g a lii  -20- ONIA - 'A3 1S56-0 -XVA ACUBWU El ....AMft.......L.. UA L -~~~~~~~~~~~I 09* n * mEAXL uUIcU I13C (1) Ocmmit, _sa im cintas (n) 1daistretie 1 &n: 27 it? 23 100 21 9t UseleT b VW 24 I1 22 100 21 o1 Ysestiam lhxaiaia NOW* 4 wntdaps 27 '117 23 100 21 91 idmy Sch*4l Module 5 clasaorn. 21 117 - 23 100 21 91 cil4-caxe NAals CAlis. 27 116 23 100 133 576 (2) Umcatlma Wg 1chool costctiom clausreome 1 100 15 100 344 210 C.steetia of eadmiistrative sa esitagY ciflsy pofsy school f_1cities fsoilieia 204 - 0 - I" - tiSOtaiom of eRitdag pdm schools clasroom 86 43 200 100 393 19f _otXaeim of ssatia sad batr,tivs millm pdriaey scool fiitis fcii 0 - 0 - 219 - Suply sA italltio of eqipint Md mteuials fez prism scbhoos /d clasoms 344 144 250 100 440 176 Trsiala of prisary teaches COMO" 132 100 132 100 190 144 (3) Reath Supply sad istalation of eqIpsout for elefh caters MAC 26 1t 22 100 23 105 Training of parmical, staff 1 staff 400 100 400 tO0 98 25 Cometctir Of eblth cener outs" cv aAC I 100 1 100 1 100 (4) iu and amil gefalse "Tl d insltllati of eqipint for CADs CAM 27 1i6 23 100 1:1 5t7 lanoatio of hom-child-cas uit _ * s 310 172 16O 100 sO 32 0ldrsa servd In me-child-oars uits Ng aduild 3,100 41 7,550 100 4,574 61 Uqaigmnat for hows-cbild-care units s0ms 310 172 ISO 100 56 32 Teaimiaof CAXP Stuag CAD. 27 IU 23 100 191 630 T"o 39 Of hildren seV" 39 OsIUn 8,100 74 lt.ODO 100 14,.56 133 UnUWDAS PRUDCUrn IIffSUDMIC Vocational Iraim ziae~ tbgldwp 27 142 19l tOO 16s COWL%" Ng 8toksto 26.100 49 52,84 100 49,1U t 31 itm tS 1,600 122 1,475 100 9o 62 Avsen Crdit Amont W 2,200 8 2,50 0 OO 3,200 In 0C iI iiC cijto Coeps * Coops 27 117 23 100 26 113 (a) Assistance to hell ntabria Group san IndIvIdal