Liberia - Monrovia Urban Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Repot No. 7865 PROJECT PERFORMANCE AUDIT REPORT LIBERIA MONROVIA URBAN DEVELOPMENT PROJECT (CREDIT 1223-LBR) JUNE 23, 1989 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. ABBUVIATIONS AND ACRONIMS ADD - African Developent Bank CDC - Comonuath Development Corporation DPD - Development Projects Department, of MCC LEDI - Liberian Bank for Development and Investent LEC - Liberia Electricity Corporation INSC - Liberia Water and Sewerage Corporation MCC - Monrovia City Corporation MLG - Ministry of Local Governwent HLAM - Ministry of Lands and Mines MUDP - Monrovia Urban Development Preparation Team NFA - National Housing Authority NHSB - National Housing and Savings Bank PPF - Project Preparation Facility USAID - United States Agency for International Development CURRENCY EQUIVALENTS The official monetary unit is the Liberian dollar with a par value equal to the US dollar. The US dollar is a legal tender in Liberia. FISCAL YEAR July 1 to June 30 WEIGHTS AND MEASURES 1 hectare (ha) = 2.47 acres I kilometre (km) = 0.62 miles 1 km2 - 0.39 miles2 FOR OFFICIAL USE ONLY THE WORLD BANK Washington. D.C. 20433 U.S.A Office of DrectoCenehual Opertms EvaluatHin June 23, 1989 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECTt Project Performance Audit Report on Liberia - Monrovia Urban Development Project (Credit 1223-LBR) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Liberia - Monrovia Urban Development Project (Credit 1223-LBR)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authoriation. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT LIBERIAs MONROVIA URBAN DEVELOPMENT PROJECT (Credit 1223-LBR) TABLE OF CONTENTS Page No. Preface ....................................................... i Basic Data Sheet .................... ........................ ii Evaluation Summary ......................................... . iv I. BACKGROUND ................................... 1 A. Economic Context ............................ 1 B. Socio-Cultural and Political Factors ................. 2 C. Institutional Context: Central Government ........... 4 D. Institutional Contexts MCC and Central-Local Government Relations .......................... 4 E. Local Service Delivery: Public Sector Performance and Private Initiatives .............. 6 F. Urban Development Tendencies and Characteristics ..... 7 II. PROJECT DESIGN ........................................... 9 A. The Project ...... ............ 9 B. Project Costs and Financing ...................... . 12 C. Unresolved Issres and Special Problems ............... 12 III. PROJECT IMPLEMENTATION ............. .................. 15 A. Overview ................... ........................ 15 B. Infrastructure ................................. 15 C. Land Adjudication ........... ...................... 17 D. Institutional Development ........................... 17 E. Project Costs and Disbursements ..................... 18 IV. PROJECT RESULTS ...................................... .... 19 A. Physical Accomplishments ............................. 19 B. Economic Reevaluation and Financial Performance ...... 20 C. Institutional Perfo7 mance .......................... 20 D. Project Sustainability ............................... 22 E. Bank Performance ......................... 23 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. V. CONCLUSIONS AND LESSONS LEARNED ........................... 24 A. Conclusions ........ ................ 24 B. Lessons Learned ............................. 26 ANNEXES 1. Total Project Costs ............ ................. ...... 29 2. Credit Categories ........................................ 30 3. Accumulated Disbursements Schedule ....................... 31 PROJECT COMPLETION REPORT Background .............................. 33 Project Objectives ....................... ..... 33 Project Description ............... .. ...... ....... 33 Project Implementation ...................... 34 Achievements and Failures .................................... 35 Lessons Learned ..................... ....... ....... 35 L PROJECT PERFORMANCE AUDIT REPORT LIBERIA: MONROVIA URBAN DEVELOPMENT PROJECT (CREDIT: 1223-LBR) PREFACE 1. This is a Project Performance Audit Report (PPAR) on the Monrovia Urban Development Project. Approved on March 11, 1982, the project was supported by Credit 1223-LBR for US $10 million, which was about 80 percent disbursed. 2. The PPAR consists of: (i) an Evaluation Summary and a Project Performance Audit Memorandum (PPAM), written by the Operations Evaluation Department; and (ii) a Project Completion Report (PCR), drafted by the staff of the Infrastructure Operations Division of the Western Africa Department (AF4IN). The PCR was distributed on June 24, 1988, about one year after loan closing. 3. The PCR provides a short account of project experience, including achievements and failures, without elaborating on the reasons for these achievements and failures. To give a wider perspective, OED has attempted to provide additional information about the history of the project, the socio- economic and political conditions encountered, the issues raised, the problems of implementation, and poscible alternative approaches. However, the Audit does not diso-gree with the conclusions reached in the PCR. 4. Because the Bank suspended relations with Liberia in 1986, it was only possible to perform a Desk Audit. However, OED interviewed most of the Bank staff associated with the project, and it gratefully acknowledges the assistance received. In preparing the PPAM, OED also consulted the President's Report, the Staff Appraisal Report (SAR), the legal documents, project files and other relevant material. PRCJECT PERFORMANCE AUDIT BASIC DATA SHEET LIBERIA: MONROVIA URBAN DEVELOPMENT PROJECT (CREDITt 1223-LBR) Key Proiect Data Original Actual or Iter Plan Reestimate Total Project Cost (US$ million) 13.337 not available Cost Overrun (Z) - - Credit Amount (US$ million) 10.0 10.00 Disbursed 7.99 Cancelled 2.01 Repaid 0.00 Outstanding as of 09/31/88 9.17 Date Physical Components Completed 6/85 6/87 Proportion Completed by Above Date (Z) 55 100 Proportion of Time Overrun - 50 Financial Performance Poor Institutional Performance - Fair Economic Rate of Return (Z) 12 not reestimated Other Project Data Original Item Plan Actual First Mention in the Files 02/79 Government Application - 06/79 Appraisal 04/81 11/81 Negotiations 06/81 08/81 Board Approval 09/81 03/82 Effectiveness 09/81 10/82 Closing Date 06/85 06/87 Borrower Republic of Liberia Executing Agencies The Monrovia Urban Development Project (MUDP) Unit within the Monrovia City Corporation (MCC). The Ministry of Local Government (MLG) Ministry of Lands snd Mines (ML&M) Liberia Water and Sewerage Corporation (LWSC) FY of the Borrower July 1 to June 30 Follow-on Project None Field Mission Data Item Month/Year No./Weeks No./Persons Staff Weeks Report Date Ident. 04/79 1.0 1 2.0 04/79 Ident. 05/79 2.0 1 2.0 06/79 Ident. 06/79 2.0 1 2.0 06/79 Prep. 09/79 1.0 1 1.0 10/79 Prep. 11/79 2.0 1 2.0 12179 Prep. 10/79 2.0 2 4.0 01/80 Prep. 01/80 2.0 2 4.0 08/80 Recon. 05/80 1.0 1 1.0 07/80 Preap. 07/80 3.0 1 3.0 09/80 Preap. 07/80 3.0 3 9.0 09/80 Preap. 08/80 4.0 4 16.0 10/80 Super. 02/81 1.0 1 1.0 12/81 Super. 01/82 1.0 1 1.0 03/82 Super. 05/82 2.0 2 4.0 06/82 Super. 09/82 3.0 1 3.0 11/82 Super. 02/83 2.0 2 4.0 04/83 Super. 02/84 1.0 1 1.0 02/84 Super. 08/84 1.0 1 1.0 10/84 Super. 04/85 1.0 1 1.0 05/85 Super. 02/86 2.0 1 2.0 03/85 iv PROJECT PERFORMANCE AUDIT rEPORT LIBERIA: MONROVIA URBAN DEVELOPMENT PROJECT (CREDIT 1223-LBR) EVALUATION SUMMARY Introduction 1. Between 1974 and 1980, the Bank disbursed nine loans and three IDA credits to Liberia, including projects to improve Monrovia's water supply and power sectors and its primary education system and small scale enterprises. The Monrovia Urban Development Project was intended to complement these projects, focusing upon the needs of the poor, as identified in a diagnostic study of the Liberian capital city carried out in 1976-78. While the project was appraised in 1979, Board presentation was delaye4 and it had to be reduced in scope because of the April 1980 military coup. On March 11, 1982, the Executive Directors of IDA approved a credit, equivalent to about US $10 million to finance roughly 76 percent of total project costs, estimated at US $13.3 million. The Project became effective on October 21, 1982. Objectives 2. This first urban project in Liberia had three major objectives: slum upgrading, land adjudication and institutional development. These concerns were interlinked, with the cost of the upgrading activities to be recovered through improved property tax collections which, in turn, required the establishment of a land ownership registry and the strengthening of municipal management. Land title registry was considered to be particularly important. Such a system would facilitate the expansion of property tax revenues. It would also p3rmit the establishment of an effective mortgage system which was necessary to overcome the obstructed flow of housing finance for lower-income groups. This component was intended to be implemented in cooperation with USAID and the Liberian Ministry of Finance. 3. Under the project, improved urban services -- including water, sanitation, roads, drainage, refuse collection, security lighting and community facilities -- were to be provided to three settlements in Monrovia (Claratown, Sonewein, and Slipway), containing a total of about 25,000 people. About 5,000 residents, expected to be displaced by planned infrastructure, were to be assisted in finding building plots and obtaining home construction loans. Improved water service, through extension of the Liberian Water and Sewerage Corporation's (LWSC) standpipe program, was expected to benefit another 40,000 people. Other components designed for parts of the city outside the three communities to be upgraded included construction of a new school, a community health center, three health units, one new market (including a small day care facility), the upgrading of an existing school, renovation and expansion of an existing market and equipment for improved solid waste management. Altogether, about 100,000 people (one fourth of Monrovia's population) were expected to benefit from the project. V Implementation Experience 4. Project execution began in June 1979 with the establishment of the Monrovia Urban Development Project (MUDP) team -- an autonomous unit within the Monrovia City Corporation (MCC). MUDP, however, was undexmined from 1982 onwards by delays in obtaining executive approval of a decree transferring 30 percent of property tax receipts to MCC. With transfers from the Central Government representing only between one half to two thirds of budgeted requirements, MCC was left with an unreliable and inadequate revenue base. Because of MCC's inadequate financial planning, budgeting and accounting systems, the local counterpart funds needed to finance capital investments and/or ensure adequate maintenance of these investments were not made available to MUDP in a timely fashion. 5. MCC never gained the power to collect property taxes in the city agreed to by the Borrower in the Credit Covenant. Even though the required decree giving MCC authority to collect such taxes was passed early Ga during the project execution perf )d, it was never effectively implemented due to strong political and administrative resistance on the part of key Central Government ministries. As a result the Ministry of Finance continued to exercise tax collecting authority in Monrovia, and it did so inefficiently. Moreover, after the 1980 coup, many landlords left the country, fearing disorder and a declining economy, thereby eroding the local tax base. MCC was further undermined when market revenues were transferred to the Liberia Market Association (LMA) despite additional costs forced on MCC by the construction of new markets and the expansion of existing ones. Revenues for garbage collection in Monrovia also dwindled, largely because of poor service. 6. The land adjudication component also ran into difficulties because it lacked the full support of the Ministry of Lands and Mines (ML&M). Because of conflicts between ML&M and MUDP, delays occurred in the provision of vehicles, equipment and supplies to the adjudicstion teams. Project funds and equipment were apparently misused and fear of property taxes may have also hampered implementation of this component. Consequently, of the 1,700 parcels of land surveyed under the project, only 241 titles were registered. After 1984, moreover, no further claims or registrations were made. 7. On the more positive side, the project's infrastructural investments were largely completed within budget, despite some delays, shortages of counterpart funds and political interference and turmoil. Except for five wetcores which could not be constructed due to the unwillingness of property owners to relinquish land for this purpose, all facilities originally planned for the settlements to be upgraded were completed and turned over to the Liberian Government by the end of 1986. The LWSC standpipe extension program was also completed during 1986. With few exceptions, the planned facilities outside the upgrading settlements were also built. However, the programmed Government housing loans for Oinfill plots* and local funds for the acquisition of refuse collection equipment were never made available. 8. Because of the incomplete implementation of the land adjudication component, land value increases resulting from the Monrovia project cannot be vi readily determined. Furthermore, while as many as 130,000 people were reported by a Liberian official to have benefitted from the operation, the Audit cannot independently assign a value ao the number of actual project beneficiaries. However, based upon the observations of Bank supervision staff, it is likely that the project did, in fact, benefit most of the residents o& the three upgraded communities. Sustainability 9. This project, which was intended to be completed by June 1985, was eventually closed at the end of 1987. In 1986 Liberia was cut off from assistance by the World Bank, the International Monetary Fund and the African Development Bank for nonpayment of its debts. At the beginning of 1987, the country was reported to be US $430 million in arrears on a $1.3 billion foreign debt. USAID also withdrew support for Liberian governmental operations in December 1988. 10. Because this is a Desk Audit, it is difficult to determine how much of the project may be sustainable. Based on its general satisfactory performance, however, MUDP revealed the potential to continue to undertake project planning, engineering, accounting and community development activities. Its Community Development Section has been notably successful in working with neighborhood associations to formulate service charges and achieve some cost recovery. Furthermore, certain local groups, such as the Claratown Development Association, have developed the capacity to take over such functions as development control, maintenance of roads and drains, sanitation and solid waste management. 11. On the other hand, the basic institutional development essential for the successful implementation and long-term sustainability of a project such as this one has not yet taken place in Liberia. Many of the problems indicated in project files (including the misallocation of funds, irresponsible Central Government interventions, fragmented governmental operations, weak financial controls and accounting systems, ministerial disinterest in urban problems and insufficient decentralization) persist according to the results of Lhe "National Seminar on Urban Development and Management Strategies* sponsored by MUDP in 1986. Central Government reluctance to transfer responsibility for the collection of property taxes to the municipality, moreover, has had a demoralizing effect on the MCC and has contributed to public resistance to pay for local services. Conclusions 12. Because the physical achievements of this project may not be sustainable and the land adjudication and institutional strengthening components generally failed to meet their objectives, the project as a whole cannot be considered a success. Nevertheless, within the context of prevailing Bank policies and procedures, the project was innovative, the proper preparatory steps -- including the use of specialized consultants and the funding of necessary feasibility studies -- were followed and a successful Project Launch was undertaken. Furthermore, Bank staff helped to establish a well-functioning project unit and, during the course of implementation, were vii able to correct problems that threatened to undermine the execution of physical works. As a result, most planned physical construction was successfully completed. 13. Yet, inasmuch as the project failed to achieve its financial and institutional objectives, questions must be raised as to what vent wrong. In addressing these aspects, the Audit suggests consideration of the following points: (a) Because the civil works -- the component of greatest interest to the Liberian Government -- were largely completed by 1984. the financial and institutional reforms specified in the Credit Agreement and recommnded by the 1984 Price Waterhouse municipal management study were zargely ignored. If the civil works could have been delayed. pending the achievement of institutional reforms, the project as a whole would have had a greater chance of success. (b) The Second Water Supply Project (Cr. 1563-LBR) for Liberia, which was appraised in February 1985, required LWSC to improve its billing and collection system, introduce a standpipe privatization plan, institute an affordable charge structure and ensure that standpipes were well maintained. Ideally, the approach followed in this project should have been introduced in the Monrovia Urban Development Project. (c) It is Bank policy to avoid intruding into domestic political matters. While the dangers of alienating the Borrower must be recognized, more open discussions about the necessary preconditions for institutional and financial reform should not be avoided even where politically sensitive. (d) It is impossible to anticipate all the opportunities and difficulties that may arise in the course of project execution. Therefore, if Bank staff had possessed greater opportunity to experiment with alternatives -- rewarding with funds those subprojects that offered the greatest potential of being self- sustaining -- the Monrovia project might have been more successful. However, such flexibility would have required more intensive and continuous Bank supervision. (e) In order to build Government commitment to project objectives and to successfully bargain for needed institutional reforms and other policy measures, the Bank needs to maintain a high level of incentives, primarily in the form of an adequate volume of funding for projects considered important by Borrowers. In the Monrovia project, the implementation experience suggests that the loan amount was clearly insufficient to motivate reform. viii Lessons Learned 14. In retrospect, it is evident that this project was considerably more risky than suggested in the Staff Appraisal and President's Reports. The preconditions for the project's full success ultimately rested upon such complex factors as political and administrative progress, a strong Government commitment to development, a meritocratic bureaucracy, public respect for the government and the existence of adequate procedures for resolving disputes, among others. Liberia's institutional and political barriers to achieving the urban development objectives desired by the Bank should have been more carefully assessed and discussed wita the Borrower. Because of the Government's traditional practice of promising -- but not necessarily enacting, reform -- an approach should have been considered which would have more effectively induced the Borrower to make institutional progress during the course of pro;ect implementation. 15. The results of the Monrovia project confirm the validity of the step-by-step approach suggested in a 1987 OED report concerning the Bank's experience with rural development. Consistent with this approach, cost- recovery arrangements could have been established prior to undertaking infrastructural investments and capital works. Project funds could have been released in tranches, based upon the attainment of sufficient progress with regard to institutional development objectives. During or after each step, the project could have been reconsidered and reevaluated. This would have allowed for contingencies not anticipated at Appraisal, while introducing greater flexibility with regard to implementation of the project as a whole. In the Liberian case, moreover, the potential of utilizing existing maighborhood groups to carry out local improvements could have been capitalized upon through establishment of a community development loan fund or a matching grant system. 16. In undertaking risky projects such as this one, the Bank needs to spend greater effort and resources on supervision if it hopes to significantly improve their chances of success. In such cases -- as an alternative to increasing or extending supervision missions -- consideration should be given to using a resident manager. A local manager could, at the same time, assume responsibility for overall supervision of the Bank's urban and related (eg. water) projects and provide more effective liaison with other donors concerned with urban development. With the right sort of person -- ie. someone concerned with community, institutional and financial development, as well as the implementation of physical works -- the Bank would have been in a better position to encourage achievement of the multiple objectives sought in this project. PROJECT PERFORMANCE AUDIT MEMORANDUM LIBERIA: MONROVIA URBAN DEVELOPMENT PROJECT (CREDIT: 1223-LBR) I. BACKGROUND A. Economic Context 1/ 1. At the time of the 1980 coup, the Liberian economy was largely dependent upon the export of iron ore and rubber, which accounted for 52 and 14 percent of export earnings, respectively. Demand for these products, however, decreased during the second half of the 1970s, causing real GDP growth to fall to less than 1 percent a year during this period. The mining industry was hardest hit, both by low iron ore prices and by the need to exploit new reserves, requiring large-scale capital investment. The export of other raw materials also declined. 2. The 1980 coup caused considerable turmoil and disorder, undermining confidence in the Government and contributing to a 4.4 percent drop in real GDP in 1980. Many expatriates and members of the local elite left the country. Foreign aid was substantially reduced. Capital outflows averaged $30 million a year between 1979 and 1981. Manufacturing -- which was largely foreign owned and accounted for about seven percent of GDP (employing about 6,000 persons) -- was also adversely affected by expatriate flight, as well as by severe power outages. 3. While about 70 percent of Liberia's work force is in agriculture, all but 10 percent of the rural labor force is estimated to be predominately engaged in subsistence activities. Little is exchanged beyond the village. Growth in agricultural output declined from a 3.3 percent annual rate in the 1960s to 2.1 percent in the 1970s. Rice production was particularly inadequate, forcing Liberia to increase imported food grains from 1,300 metric tons in 1978 to 26,300 metric tons by 1981. 4. Parallel to the decline in exports and GDP during the 1970s, Liberia's budget deficits increased from roughly US $4.0 million in FY 1975 to US $161 million by FY 1979. Annual budget deficits during the early 1980s continued to average around US $100 million. The foreign account deficit rose to about US $130 million in 1979, amounting to 13 percent of GDP, as compared with US $30 million in 1974, or 5 percent of GDP. Between 1970 and 1980, Liberia's public debt (from both official and private sources) increased from 1/ The sources for this section includes The President's Report for the Project March 11, 1982; J. Gus Liebenow, Liberia: The Quest for Democracy (Bloomington and Indianapolis: Indian University Press, 1987); and Chris Hermann, Margaret Shaw, and John Hanna, Development Management in Africa: The Case of the Agricultural Analysis and Planning Project in Liberia (Washington, D.C.: USAID, December 1985). 2 US $158 million to US $640 million. By 1987, Liberia was US $430 million in arrears on a US $1.3 billion foreign debt. 5. The President's Report for the project suggests that Liberia has had a long history of sound fiscal management. This judgment is doubtful, however, for the Tolbert era (1971-1980) when the Government supported the establishment of a number of unprofitable enterprises. Foremost among these were the Presidential city of Bentol (with its lavish villas, private zoo, artificial lake and public buildings), the new university campus at Fendel, the new airport terminal building and the lighting of the 60-kilometer road from Roberts Field to Monrovia. It is estimated, moreover, that the Tolbert administration speut over US $200 million to host the 1979 summit meeting of the Organization of African Unity, when it built a resort hotel and more than 50 chalets to house visiting heads of state. In contrast, at the same time the Government neglected to maintain existing sever and drainage systems in the central parts of Monrovia and paid little attention to the poorer parts of ehe capital city. 6. The Doe regime, which came to power in 1980, further strained public finances by sharply increasing the salaries of many military and civilian public sector employees. The Liberian Government, however, subsequently attempted to control public sector expenditures under a series of Standby Agreements with the IMF. Decreases in rice price supports, a hiring freeze, reduced official travel expenditures, a reduction of salaries by 16- 25%, tighter fiscal policies and stricter control over the financial operations of public corporations resulted. 7. By 1983, development projects were being funded largely by external donors. However, administrative mismanagement, evasion of controls and extrabudgetary expenditures undermined the Standby Agreements. Between 1979 and 1985, there was also a six-fold increase in military expenditure. Despite continued USAID financial support and technical assistance until December 1988 -- when it removed its financial advisors from the country and confined its aid to local NGOs -- the Liberian economy has continued to deteriorate. T'e recent substitution of newly minted Liberian coins for American paper currency has further undermined confidence in the economy and destabilized commercial transactions. B. Socio-Cultural and Political Factors 2/ 8. Social and cultural factors have affected Liberia's economy in various ways. Subsequent to the colonization of the country by ex-slaves from the United States in the 19th Century, it was dominated by their descendants 2/ In addition to the sources cited earlier, the following have been used for this sections Harold D. Nelson, ed. Liberia: A Country Study (Washington D.C., U.S. Government Printing Office, 1985); Anthony J. Nimley, The Liberian Bureaucracy: An Analysis and Evaluation of the Environment, Structure and Functions (Washington D.C., University Press of America, 1977); James Brooke, "US Will Oversee Liberian Finances," The New York Times, April 26, 1987. 3 (referred to as "Americo-Liberians") until the 1980 coup. As a result, an estimated 5 percent of the population controlled the vast majority who were perceived to be or considered themselves as "tribal." As a result, members of Liberia's 17 native ethnic groups were largely excluded from the political system. 9. Until 1980, tribal people were subject to compulsory labor, porter service and the jurisdiction of tribal courts. An estimated 50 percent of their cash assets were consumed by a hut tax. While tribal people did manage to gain an increasing number of Government positions, the highest posts were reserved for members of the Ameri.to-Liberian elite. Educational opportunities were also largely limited to this group. While the elite was relatively well- educated, however, few were trained or interested in areas such as agronomy, engineering, and business administration. At the same time, 75 percent of the Liberian population remained illit_erate, and only 50 percent of all children were enrolled in school. 10. A 1966 Northwestern University study found that 75 percent of the national income went to foreign households and business firms or to the 3 percent of the Liberian population which formed the political elite. 3/ The study concluded that much of the growth that occurred during the post-World War II years had not led to meaningful development. In 1970, it was estimated that 40 percent of all wages, salaries and benefits went to 50,000 expatriates in the foreign concessiors, much of which left the country in the form of repatriated salaries. While some progress was made by Liberia's tribal people under the Tolbert regime, the Ministry of Planning and Economic Affairs pointed out on the eve of the 1980 co-ip that only 4 percent of the population owned more than 60 percent of the national wealth. 11. During the Tubman (1944-1971) and Tolbert periods, an effective road and bridge network was built throughout the country. Rail links were added to connect the coast with important mining areas. While these transportation improvements encouraged tribal peoples to enter commercial agriculture, they also increased their exposure to exploitation by the Americo-Liberian elite. The elite used its education, economic resources and political power to take control of large sections of tribal land, in the process undermining traditional agricultural and social systems. Consequently, many tribal people left agriculture for urban areas, particularly Monrovia. In doing so, they contributed to worsening conditions in the capital city inasmuch as the Government generally failed to meet their basic needs, especially for inexpensive food. 12. The coming to power in 1980 of Master Sergeant Doe and his colleagues, who called themselves the People's Redemption Council (PRC), was initially perceived as a revolution, frightening the Americo-Liberian elite, while elating the tribal people and supporters of .3form. However, to minimize anxiety and chaos, the new Government restored much of the property initially seized from foreigners and the local elite. As time went on, the 3/ Robert W. Clower, et. al., Growth without Development: An Economic Analysis, 1966. 4 military nature of the PRC became clearer. While respected civilians were given top positions in most ministries, real power rested with members of the armed forces. By mid-1984 the initial sense of mission and optimism that many Liberians felt after the coup had evaporated. While certain ethnic groups with ties to the new military rulers gained ground at the expense of the Americo-Liberians, the standard of living of a large proportion of the population declined. C. Institutional Context: Central Government 13. The Liberian administrative system is a prototype of "patrimonialism." Family relationships were particularly important prior to the 1980 coup. Patronage was used to maintain a balance of power among the leading families. This system, however, was ultimately counterproductive, undermining the mutual respect among officials essential for cooperation and cohesive public administration. Senior Liberian advisors were frequently dismissed or rotated, and excessive reliance tended to be placed on foreign ,dvisors. Moreover, because of inadequate planning, record-keeping and statistical information, priorities were seldom established or projects completed. 14. Various efforts over the years were made to reform this system. In 1973 the Liberian Institute of Public Administration was established for this purpose with the USAID-funded assistance of the New York Institute of Public Administration and the United Nations Economic Commission for Africa. These reform efforts were largely unsuccessful. A USAID study during the early 1980s found the persistence of administrative problems, including personalized and ad hoc decision-making, informal personnel appointments and a general lack of accountability for official actions. 15. Formalism continues to characterize all aspects of Liberian political and administrativ- affairs, particularly in the legislature, the courts and the election system. While the Constitution of 1847 (which was modelled on the American charter) was given extensive homage prior to 1980, it was never made fully effective. In reality, the President -- through his domination of the one-party system, the army, and the police -- maintained near dictatorial control over political developments in the country. D. Institutional Context: MCC and Central-Local Government Relations 16. Although the city of Monrovia gained the status of a corporate body with independent operating powers in 1911, under a 1922 Act it remained fully under the control of the Central Government. The corporate status of Monrovia was officially restored by the 1973 Charter, which set out the organization and functions of the Monrovia City Corporation (MCC). However, except for the formal creation of the MCC, the provisions of the 1973 Charter were never put into effect. 17. While MCC was given further powers in 1976, actual responsibility for most local government functions continued to rest with the Ministry of Public Works. As the Staff Appraisal Report recognized, most Central Government ministries were very amb-1-yalent about decentralization. Because 5 few central agencies were inclined to share their authority, staff or budget, the expansion of MCC's duties has been slow and hard-won. With the suspension of the National Constitution in 1980, the status of the 1973 Charter became uncertain. Although never formally repealed, many of the Charter's provisions were not implemented. 18. The refusal of the Liberian government to give greater autonomy to MCC had a number of negative consequences. In 1980-81, for example, MCC was able to directly collect only about 7% (US $164,000) of its budget expenditures (estimated to be US $2.2 million). These revenues were restricted largely to fees from the cemetery, petty traders, garbage collection and markets. Despite a rather elaborate administrative structure, moreover, MCC had no spending authority and little to administer. 19. In November, 1979, Monrovia's first Mayoral election was scheduled to be held. A leader of the country's major reform party, the Movement for Justice in Africa (MOJA), made a bid for this position in August of that year. The Central Government, fearful of a MOJA victory, however, postponed local elections until June of the following year. With the change, in regimes in April 1980, the municipal elections were cancelled. As a result, MCC was left with a series of Acting Mayors appointed by the President. While some of the Acting Mayors were competent and motivated, they, nevertheless, lacked an independent political mandate. Furthermore, Mayors and other city officers were frequently dismissed or rotated. With each change of local authorities, Monrovia's population became increasingly skeptical about Central Government promises to give additional functions, staff and budgetary authority to the City. 20. Under the Doe regime, the duties of the Mayor and MCC were never specifically defined. Functions tended to be transferred on an ad hoc basis, and there wcre many areas of overlap and dupltcation, giving rise to inefficiency in the use of manpower and other resources. 4/ The City's activities and operations needed to be approved by the President or his deputy, and these approvals were often difficult to obtain. 21. To further complicate matters, different ministries took independent actions affecting MCC, with little effective coordination and no adherence to a set of priorities. USAID found, for example, that its efforts to draft a national housing policy and plan in 1979-80 failed largely because of a lack of interest among top Government officials. Without authority, many of the best qualified professionals left city employment. Other, fearful of being detained or dismissed, left inexperienced junior officers in charge. For much the same reasons, newcomers were reluctant to take decisions and refused to exercise leadership or to take initiatives. In part as a result, financial statements were seldom prepared by MCC on funds spent under its control, no satisfactory accounting system or audit arrangemc s were 4/ Price Waterhouse Public Accountants, Monrovia Urban Development Project: Municipal Management Study, Final Report, Vol. I., July 1984. 6 established, even office supplies were often lacking and various forms of mismanagement occurred. 5/ E. Local Service Delivery: Public Sector Performance and Private Initiatives 22. The above mentioned local government deficiencies had a direct impact on the delivery of public services in Monrovia. Typical of the problems encountered were those experienced in the water sector. Water bills varied sharply from month to month, meter readers and workmen occasionally asked for bribes, consumers received discourteous service and water supply was often arbitrarily disconnected. As a result, many people made illegal or unauthorized connections to the city water system. In 1980, it was reported that the Liberian Water Supply Company was billing about 50% of the water produced. Of the remaining 50%, half was lost from leakage and the other half was lost by illegal connections and underbilling. 23. In the electrical energy sector, the situation was largely similar. A cat and mouse game of disconnections and illegal reconnections was carried on, often with the assistance of former employees of the Liberis Electricity Corporation (LEC). Large, as well as small, users avoided paying bills, often wasting substantial amounts of electricity in the process. Since records were poorly kept, LEC could do little about this situation. 24. In other areas, however, the experience was more positive. An internationally respected group of individuals, many of whom were associated with the University of Liberia and the MOJA, were able to challenge the Government both before and after the 1980 coup. At various times-- especially after the coup -- some of these individuals held ministerial positions, even though their real authority remained limited. Nevertheless, projects which they undertook -- particularly for the development of rural marketing cooperatives -- received various forms of foreign assistance and were often successful, despite governmental interference. 25. In Monrovia, moreover, local chiefs often took a genuine interest in community affairs, organizing households to pay for improvements such as road and footpath extensions, drainage, health and day-care services. Churches, ethnic associations and mutual-aid groups were also very active. Many people were interested in land registration and hired lawyers to assist them without response from the Government. A 1980 consultant's report on urban poverty in Monrovia found that even poor people were willing to pay the estimated costs of water, electricity and other basic services, if the administration of these services could be improved. A year earlier, another consultant reported that people living in low and middle income settlements in the capital city had formed Community Health Motivation Councils to improve neighborhood drainage systems and build new public toilets and dump sites to facilitate garbage collection. 5/ L. Victor Smith, Liberia - Urban Development I - Preappraisal, September 2, 1980. 7 F. Urban Development Tendencies and Characteristics 6/ 26. A 1977 USAID Shelter Sector Analysis estimated urban population to be growing at 8-10 percent annually. By 1980, nearly a third of Liberia's 1.9 million inhabitants lived in urban areas, of which about 50 percent were in the capital. Between 1950 and 1980, Monrovia's population quadrupled, reaching about 400,000. Monrovia was expected to add another 340,000 people by 1990, thereby nearly doubling its population during the 1980s. The opening of new roads and the improvement of transportation and communication facilities, the expansion of educational and economic opportunities, the introduction of the money economy into remote areas and the increasing resentment by tribal populations of Americo-Liberian land grabbing and rural control, all contributed to the rapid growth of Monrovia. 27. Liberia's high overall population growth rate (3.5 percent) has also contributed to Monrovia's 7 percent annual growth rate. By 1980, the number of tribal people ir.tegrated in the money economy was estimated to have reached 50 percent, as against about 25 percent in 1970. Urbanization was spurred by the fact that one-third of the adult male population -- a higher percentage than in most African countries -- were in relatively high-income wage employment, according to a 1974 census survey. Such secondary cities as Sanniquellie and Buchanan also grew rapidly between 1950 and 1980, primarily as a result of the iron industry. However, compared to Monrovia, these cities remained small and were adversely affected by the decline of the iron industry. Buchanan, the second largest town at 20,000 was only one-fifteenth the size of Monrovia in 1980, and the next largest towns had less than 10,000 inhabitants each. 28. In 1980, it was estimated that 40 percent of Monrovia's population lived in areas needing extensive upgrading. Because of the confusing land ownership situation and lack of control over land use, some 38 percent of all households in Monrovia were legally squatters. While squatter areas had some access to electricity and piped water, service was unreliable, particularly in the case of water. Public standpipes were frequently broken and water pressure was inadequate and uneven, forcing people to buy water at exorbitant rates or to rely upon roof cisterns or contaminated shallow wells. The cost of potable water -- about 8 cents per gallon, or about fifty times the Liberian Water and Sewer Corporation's official tariff rate -- was so high that the average household utilized only about six gallons per day, mostly for drinking and cooking. 29. Street or security lighting was almost universally absent, contributing to high crime rates in many areas. The existing sewer network was inadequately maintained, eve . in the city's central area, and did not reach poorer areas. The lack of t..orm drainage worsened sanitary conditions 6/ In addition to the ccnsultants' reports and the Staff Appraisal Report on Monrovia cited earlier, use has been made of Liberia: Urban Project Brief, February 17, 1980 and the National Seminar on Urban Development & Management Strategies (Monrovia, Liberia, MUDP, June 1896). 8 and exacerbated flooding during Monrovia's frequently heavy rains (over 200 inches a year). Inadequate garbage collection further contributed to the problem of poor environmental sanitation, particularly in low-income neighborhoods. 30. The ccnsequences of these poor living conditions in large parts of Monrovia were seere health problems, including a high incidence of hepatitis and the frequent occurrence of endemic diarrheal diseases such as cholera, dysentery and typhoid. The mortality rate for children under 5 was extremely high, caused by a combination of water-borne diseases and malnutrition, which was estimated to affect 15 percent of this age group. Liberia's social and economic problems mentioned earlier contributed to the high incidence of urban malnutrition. In 1979, 25 percent of the rice consumed in Monrovia had to be imported. As a result, the price of a bag of rice rose to US$30 at a time when the average monthly income of urban Liberians was only on the order of US $80. This led directly to the "rice riots" of April 14, 1979 which undermined the Tolbert regime. 31. The Tolbert administration constructed five new housing projects, but only one of these was made available to low-income families. The others were allocated to middle-income occupants. Some of these houses were purchased by the elite for profitable rental purposes. In an attempt to invert this tendency, a US $15 million USAID Housing Guarantee Loan was approved in June 1979 for the development of low cost shelter solutions in two communities in Monrovia. The USAID project, which initiated in a neighborhood called New Georgia, primarily involved the development 1i sites and services, with only 26 percent of the total funding (US $4 million) utilized for squatter upgrading. The USAID-financed housing units, however, turned out to be well above the level affordable to most families in Monrovia, in the absence of heavy subsidies. 32. During the 1980-85 period, it was estimated that, of the 19,000 new households formed in Monrovia, 70 percent would have to solve their shelter problems by themselves, with practically no help from the Government or the "formal" housing sector. This meant that home construction frequently took long periods of time, with many houses remaining unfinished. The absence of land use and zoning regulations or building codes, in turn, meant that there was virtually no control over the quality of housing construction. As a result, an estimated 60 percent of all shelter solutions were built of temporary materials. 33. While many of Monrovia's residents paid rent to houseowners, they could never be certain about the actual ownership of land on which they resided. The lack of clear land tenure information in Monrovia was seen by Bank staff as a primary obstacle to orderly urban development. Without a clear definition of land ownership, rights of way could not be bought, property improvement encouraged, mortgage loans extended or investment and service costs adequately recovered. Even in the central parts of the city, moreover, disputes over landownership contributed to the poor quality of construction and to the many incomplete structures. 9 34. Without formal street numbering and an up-to-date postal box system, property owners were difficult to identify. Furthermore, since urban land sales were rare or unannounced, it was impossible to determine market values and to enforce payment of property taxes. While the law provided for confiscation and auctioning of properties for non-payment of taxes, there was no record of any property being foreclosed upon and sold. Only about 20 percent of registered property owners paid property taxes. 35. The urban employment situation in Liberia was also precarious. In May 1979, a Bank consultant carried out a study of informal sector employment. Underemployment in Greater Monrovia was found to affect about 60 percent of the labor force and unemployment, nearly 40 percent. Estimates of the formal labor force ranged from 25 to 33 percent of the total, amounting to some 60,000 jobs, of which Government and public corporations accounted for over half. About 18,000 persons were believed to earn their livelihoods through the informal sector, more than half (10,000) of whom were market women. 36. Finally, because of the economic difficulties in the late 1970s mentioned earlier, real incomes of the urban population were declining. About 50 to 70 percent of the residents in low income settlements fell below the absolute poverty line of US $310 per capita. In response to this worsening situation, USAID approved a loan for US $1 million to the informal sector through the National Housing and Savings Bank (NHSB) in June 1979 and a parallel Bank IFC project was proposed to reach small-scale borrowers. Moreover, the third Bank-assisted education project attempted to promote vocational training, and several piivate and religious organizations sponsored employment-related training schemes and small production workshops. The Monrovia Urban Development Project was similarly designed to help alleviate growing problems of urban poverty and inadequate local service delivery, as well as to strengthen local government administration and municipal revenue generation. II. PROJECT DESIGN A. The Project 71/ 37. Preparation of the Monrovia Urban Development Project was based upon the Monrovia Urban Study, financed by the Bank and carried out in 1976- 78. While the preappraisal work was completed in 1979, the Staff Appraisal Report did not appear until November 1981 because of the military coup of April 1980. The Project went to the Board on March 11, 1982 and the Credit Agreement was signed on April 21, 1982. While the project, in effect, began in June 1979, with the establishment of the Monrovia Urban Development Preparation (MUDP) team - an autonomous unit within the Monrovia City 7/ Unless otherwise indicated, the sources for this section include the Staff Appraisal Report for the project, November 30, 1918; the President's Report to IDA, March 11, 1982; the Development Credit Agreement, April 21, 1982; and the Project Completion Report, June 24, 1988. 10 Corporation (MCC) - the Bank credit did not become effective until more than three years later, on October 21, 1982. A successful Project Launch Seminar was tonducted in June 1982, attended by 200 participants, diplomats and guests, and hosted by the Mayor. 38. The initial Bank identification mission, dated April 9, 1979, listed the following possible project components: upgrading, serviced sites, small business development, urban transportation, solid waste management, secondary town improvement and technical assistance to the newly established MCC. While studies were eventually carried out to determine the feasibility of most of these components, by the end of 1980 the project was limited to three sets of interdependent measures for: (i) infrastructural improvement, (ii) land adjudication and (iii) institutional reform. 39. It was expected that through project-related training, MUDP staff would gain the skills required to manage physical implementation and maintenance of infrastructural and community improvements. Because the revenue base for 85 percent of project cost recovery and to sustain urban development activities was to come from city-wide property tax charges, the establishment of a land title registry in Monrovia was considered essential. MCC was to collect the property tax and retain 30 percent of the total beyond collection costs. Under the project, MCC would also be responsible for maintaining accounts, controlling expenditures, coordinating investments, undertaking urban planning and assisting the communities within Monrovia. All of these had heretofore been functions of the Central Government. The project was thus seen as a first step towards the objective of long-term institutional strengthening of the municipal level. 40. The infrastructure component of the project, as appraised, included water, sanitation, roads, drainage, refuse collection, security lighting and community facilities for three important low-income squatter settlements in Monrovia: Claratown, Sonewein, and Slipway, containing a total of about 25,000 people. These settlements were selected partly on the basis of their large populations and poor living conditions and partly because of their interest in upgrading and expreised willingness to repay the cost of the services to be provided. Land tenure problems and existing USAID investment programs were also taken into consideration in the selection of these areas. 41. The need to resettle households displaced by planned infrastructure works was expected to affect less than 20 percent of all structures in the settlements to be upgraded. Residents of these structures (about 5,000 people, including renters) were to be assisted in finding building plots in Claratown or New Georgia, for which home construction loans were to be made available. Compensation was to be paid homeowners for demolished structures. Expropriated land would be compensated if clear title could be demonstrated. 42. In the three settlements chosen for upgrading, home ownership and room tenancies were to be left undisturbed. No attempt would be made to alter the absentee landlord character of existing rental arrangements. However, to avoid sharp rent increases arising from project improvements, only the most basic items were included and design standards were to be kept low. Major access roads were to be upgraded in all three project settlements, with drains 11 built along one side of them. About 31 centrally located toilet blocks-- each containing 12 toilet units, shower facilities and nearby water standpipes -- were also to be built. Security lighting was to be provided in the selected settlements. Finally, a community center was to be constructed in each upgraded area to provide space for neighborhood activities. 43. About 40,000 people outside the threz neighborhoods selected for upgrading were also to be benefitted through the provision of improved water services by extension of the Liberian Water and Sewer Corporation's (LWSC) existing standpipe program. Other project physical components covering different parts of the city included construction of a new school, a community health center, three health units, one new market (including a small day care facility), the upgrading of an existing school and renovation and expansion of an existing market. Refuse collection points were to be provided along major roads, and collection vehicles, loaders and maintenance equipment were to be acquired. 44. The land adjudication componen\' was intende,A to facilitate implementation of the Land Registration Act initiated in 1974 by the Ministry of Lands and Mines (ML&M) with UNDP, and later, USAID assistance. Land registration was intended to replace the former deed arrangements with a guaranteed title system legally defining plots and the ownership of urban land. The adjudication process involved two tasks, carrying out cadastral surveys to establish plot boundaries and the subsequent registration of owners. 45. Under this component of the project, essential office and field equipment, salaries, operating expenditures and technical assistance were to be financed in support of the seven teams of the ML&M unit involved in the adjudication program. The component would be implemented in two stages. The first phase of the Land Registration Program -- which already had begun in 1979, as part of a USAID project in the West Point neighborhood -- was intended to establish land ownership in the three proposed project upgrading areas. This work was expected to be completed by the end of 1982. A second phase would the aim at strengthening and expediting land adjudication throughout Monrovia. 46. Under the institutional development component, in turn, various forms of technical assistance were to be provided to the MCC, the Ministry of Local Government (MLG), the National Housing Authority (NHA) and the ML&M. Liberian or West African institutions were to be used primarily, supplemented by courses taught by foreign experts under contract. Selected staff of the beneficiary agencies would also attend short courses abroad. Altogether, approximately 74 persons were expected to participate in courses on urban growth, traffic management, public administration, land registration, local finance and public health. 47. A number of studies were also to be financed under this component, the most important being a Municipal Management Study, which was expected to take six months. This study would provide a general review of MCC's operations, including its existing organizational structure and management, revenue accounting systems, tax collection operation and staff development 12 programs. Under the project, technical assistance would also be provided to implement the study's recommendations. B. Project Costs and Financing 48. The estimated total cost of the project was US $13.3 million, including projected domestic and foreign price contingencies and land costs. (See Annex 1 for a breakdown of estimated project costs by component.) Project cost estimates were based on preliminary engineering studies prepared by MUDP-- using rates and prices obtained from local contractors and the Ministry of Public Works -- and the Bank's experience with its Second Education Project. Nearly half of the project's total cost was estimated to require foreign exchange (US$ 6.8 million.) 49. IDA would finance about US $10.0 (SDR 8.3 million) -- or about 75 percent of total project costs -- covering the entire foreign exchange cost of the project and 49% of the lccal costs, excluding taxes and duties. The Government would be the borrower and assume the foreign exchange risk. Disbursements for civil works, consultancy services and training and all major equipment (costing an estimated US $8.5 million) were to be made against fully documented contracts and receipts. Spending authority was to be vested with MCC, using a revolving fund of US $400,000, initially composed of an IDA Special Account of $300,000 and a Central Government contribution of $100,000, the latter being a condition of credit effectiveness. The Liberian Government would on-lend funds for the project to MCC and LWSC under the same terms received from the Bank. C. Unresolved Issues and Special Problems 50. In preparing the Monrovia Urban Development Project, Bank staff expressed concern with a number of issues. A review of project files indicates that, while some of these issues were studied by consultants and discussed extensively in departmental meetings, they remained largely unresolved. As a result, Bank staff later responsible for the supervision of project implementation were faced with some recurrent and foreseeable problems. 51. Project Scope and Local Interest Bank justification for limiting the scope of the project and reducing settlements to be upgraded from four, as originally proposed, to three resulted from the uncertain political and economic conditions that followed the 1980 coup. These decisions were made exclusively by Bank staff, without input from local authorities or taking into consideration the desires of potential project beneficiaries. This meant that the necessary support of Monrovia's leaders and citizen associations was not sufficiently mobilized in advance of project implementation. During the Project Launch Workshop, it was dis,.overed, for example, that, while the Mayor received 25 calls from local groups requesting upgrading activities in their areas, the largest and most complex of the areas preselected to be upgraded (Sonewein) did not even send a representative. It was also found that, prior to the Project Launch Workshop, almost none of the representatives of the beneficiary communities were aware of the project's origins or purpose. 13 52. Land Adiudication The need for an adequate registry of land ownership was recognized early in the development of the project. Based upon visits of Bank staff to Monrovia in 1979, a Liberian Land Tenure Seminar was held at the Bank in February 1980. A number of questions were raised at this time includingt (a) the attitude of the African Methodist Church concerning the upgrading of housing units on its own land; (b) the extent tu which tribal chiefs have powers to allocate land rights of their people and collect charges; (c) the extent of opposition by large land owners and the Church to the requirement that they must pay outstanding property taxes to clear titles and register land; (d) the impact of measures to clear titles on low-income residents; and (e) the likely degree of cooperation of the courts and various governmental agen3ies. Because these questions remained unanswered, Bank staff supervising project implementation were faced with the following problems: how to most expeditiously obtain the land needed for the project, how to use property taxes to recover some of the costs of the upgrading program, how to deal with absentee landlords and how to provide secure tenure for present occupants of land (homeowners and subletters). 53. The difficulty of these issues led Bank staff to place hope In a "quick-take" provision found in Liberian law enabling services and community facilities to be provided without significant delay, even when disputes arose as to land ownership. A rent tribunal was suggestid in order to resolve disputes arising when tenants claimed that rents were rising more rapidly than property taxes. However, misgivings continued to be expressed regarding possible outside interference which would delay the adjudication process and lead to intimidation. Finally, since existing collections of property taxes were low (less than 20% at the end of 1980), opposition to the proposed property tax concept by powerful groups was also anticipated. 54. Cost Recovery Because direct cost-recovery from project beneficiaries was to be attempted for the first time in Liberia's history, it was expected to be both controversial and difficult to implement. Under the preliminary cost recovery formula discussed with the Government in September 1980, about 69% of total project costs would be recovered through direct beneficiary or user charges, utility tariffs or property taxes, and the remaining 31% would be borne by Government. Monthly charges -- covering capital costs and utility payments for recurrent services -- were expected to total US $10.65 and be affordable to about 85% of all households in the areas to be serviced. A surcharge was included to cover the remaining 15% who were likely to default. Repayment was to be at a 13% interest rate (12% plus 1% default surcharge) over a period of 25 years. However, as noted by Bank staff, after the 1980 coup it was increasingly difficult to get the new officials to focus on such financial issues as the division of costs between Government and beneficiaries, the legal basis for collecting costs from beneficiaries, the sanctions to be applied in the event of non-payment and interest rate and repayment terms to households for home improvement loans. 55. While the costs of on-site infrastructure, administration and interest during construction was to be recovered through the introduction of property taxes in the upgraded areas, the cost of water taken from public standpipes would be recovered through a charge of $1.68 per 1,000 gallons of kiosk water. Bank staff were anxious to promote metering for all water users 14 to encourage thrift and facilitate bill collection since, heretofore, water bills were infrequently paid. With regard to collection arrangements, it was recognized that there would need to be some experimentation, including the possible use of community contracts, municipal franchises for water kiosk sales and an amalgamated charge for both capital and recurrent services. 56. A franchise program was suggested whereby contractors would undertake responsibility for maintenance of the shared toilet blocks and water points in exchange for the profits on water sales. If this experiment failed, however, the Government would be required to meet the costs of water consumed at standpipes. A community collection system for property taxes was also considered. Unanswered questions regarding the best approaches for metering, determining tariff rates, billing and collection procedures were left to be resolved during the course of project implementation. 57. Decentralization Very early in the preparation of the Monrovia project, it was recognized that: (a) MCC was very weak in terms of qualified staff; (b) its responsibilities and status were poorly defined; and (c) the Central Government's willingness to strengthen the municipal agency was unclear. It was also recognized that a strong, independent executing unit within the MCC would be essential in order to obtain coordinated action from the different agencies responsible for the provision of infrastructure and services. 58. While the Central Government, in principle, expressed its support of some decentralization, several powerful Ministries feared the associated loss of budget and influence. The fact that the Acting Mayor lacked a political mandate was also an impediment to effective decentralization. Despite the recognized weakness of the MCC, it w , nevertheless, selected to implement much of the project, rather than ..e National Housing Authority (NHA) which was concurrently undertaking an USAID-financed housing project. The justification for this decision was, first of all, that this was primarily an upgrading, as opposed to a housing project, and secondly, that the Government had expressed some interest in expanding MCC's management capacity. In this regard, there appeared to be a gensral consensus about the need for a Municipal Management Study. 59. Despite the emphasis placed by Bank staff on decentralization, serious doubts remained as to how much real autonomy would be given to MCC and MUDP for implementing the project. The Mayor of Monrovia, in a 1981 memo, expressed his fear that the Ministry of Local Government would attempt to assume responsibilities assigned to the City Government. Leading members of the Central Government, on the other hand, indicated their concern that the proposed legislation for Monrovia might become a precedent for secondary cities and other municipalities in the country. 60. Despite these unresolved issues, Bank staff felt the need to go ahead with the project before all the necessary information could be obtained, institutional conditions established and managerial problems resolved. Already, by the end of June, 1979, 300 employees had been hired by MCC in anticipation of the project, and 15 people had been assigned to MUDP. Further 15 delays in initiating the project, it was argued out in July 1980, would result in a significant loss of the momentum. 61. It was hoped that the outstanding issues could be successfully handled in the process of project implementation. This later proved to have been a hazardous way of proceeding, even though it was consistent with Bank project cycle procedures at the time the operation was prepared and appraised. In retrospect, it is clear that, if execution of the infrastructure component could have been delayed until the cost-recovery mechanisms were in place, implementation of the land adjudication and institutional development components might have had a bette- chance of success. In doveloping a working relationship between MUDP and Morrovia's various communities, the selection of communities, the choice of components, and the willingness to pay for costs could have been tested, rather than prescribed. Furthermore, the project might have been made more attractive to the Borrower, with the prospect of potentially including more substantial funds and additional components without endangering the achievement of its multiple objectives. III. PROJECT IMPLEMENTATION A. Overview 62. With regard to implementation, the three basic components (infrastructural improvement, land adjudication and institutional reform) of the Monrovia Urban Development Project can be examined separately. As will be further explored below, the fact that these components were not implemented in an integrated manner makes it doubtful that the project is sustainable. Despite implementation difficulties and delays (indicated in Annex 2), the infrastructural investments were largely completed by the end of 1984. On the other hand, very little progress was made with regard to land adjudication and institutional reform. Because the civil works and public facilities (Annex 3) were being implemented within the prescribed budget, Bank staff responsible for supervising this project were reluctant to discontinue it, even though the other components were clearly unsuccessful. Largely as a consequence, with completion of the upgrading component and in the absence of a follow-on operation, Bank staff lost all leverage with the Liberian Government with respect to the other project components. B. Infrastructure 63. At the time of the first supervision mis3sion for the project (September/October, 1982), Bank staff found consultants' cost figures for the proposed civil works and community facilities far in excess (ie. by approximately US $2.8 million) of Appraisal estimates. The discrepancies were due largely to higher standards, increased scope of work, questionable cost estimates and costly design solutions. The consultants provided no briefs explaining design criteria or justifying higher standards, and preliminary designs had been approved in the different Government agencies and subsequently by MUDP without being accompanied by detailed cost estimates. As a result, a number of measures aimed at reducing costs were proposed and tentatively agreed upon in order to bring design standards and costs in line 16 with those originally agreed upon at Appraisal and Negotiations. Nevertheless, when Bank staff returned to Monrovia in February 1983, they again found a number of flaws and gaps in the final designs and tender documents, including insufficient cost estimates and excessively costly solutions. 64. In their next supervision mission (August 1983), Bank staff found that detailed designs and contract documents for all civil works and public facilities had been completed and approved by the Ministry of Public Works, Liberia Water & Sewer Corporation and MCC. Tender bids were expected to be received in October 1983, construction of public facilities to start in November and that of civil works in December 1983. The principal remaining problem was the relatively high estimated compensation costs for displaced residents at the three sites of Claratown, Slipway and Sonewein ($1.5 million, rather than the $282,000 estimated at project appraisal). Various proposals were made to reassess and reduce these costs. 65. In October 1984, Bank staff reported that -- despite severe financial problems -- physical works at all three sites were progressing well and were expected to be completed by the end of the first quarter of 1985, provided adequate counterpart funds were made available. Delays in counterpart funding, combined with a temporary freeze on World Bank disbursements, continued to cause delays. While contractors had been paid US $5.25 million for their foreign expenditures, as of April 1987, they were still owed US $1 million by the Government. Nevertheless, the MUDP Construction Progress Report for October-November 1985 estimated that the civil works and public facilities were about 85 percent complete. On the other hand, the December 1985 MUDP Construction Progress Report indicated that the completed buildings could not be used because the required sewage connection was not functioning. Consequently, some of the buildings constructed under the project (particularly the health unit in Sonewein) had been broken into and abused. 66. In their final supervision mission (February 1986), Bank staff reported that infrastructure works within the settlements had been virtually completed. The sewer line through Sonewein had been cleaned out, and all wet cores, primary schools, community centers and health units had been finished. The soccer playing field in Slipway, on reclaimed land, was proving especially popular. According to the MUDP Annual Report for 1986, all facilities planned for the settlements were concluded except for five wetcores (3 in Slipway, 2 in Sonewein) which could not be constructed due to the unwillingness of property owners to relinquish land for this purpose. The completed facilities were turned over to the Government of Liberia at the end of 1986 for use by the target population. The LWSC standpipe extension program was also completed during 1986 and included a total of 21 standpipes installed in poor neighborhoods of Monrovia outside the three upgraded settlements. 67. Some of the upgrading subcomponents (categorized as "other" in Annex 1) were not undertaken for various reasons. Housing loans for "infill plots" were never made available because of the difficult credit conditions required by the National Housing and Savings Bank. A health center for New Kru Town proved to be unnecessary. Funds for refuse equipment and disposal were 17 not made available because of lack of interest on the part of MCC in an offer of technical assistance from officials of Dayton, Ohio to improve solid waste management under a "twinning arrangement." MCC eventually used its own funds for refuse-collection equipment that proved unsatisfactory. C. Land Adjudication 68. By December 1981, the land adjudication component was already in trouble. At that time, it was reported that the Ministry of La..-ds and Mines (ML&M) had not yet equipped the adjudication teams with the vehicles, equipment and supplies needed to carry out their work. Misunderstandings about the availability of Project preparation funds and a lack of coordination between the ML&M and MUDP apparently caused the delay. In their November 1982 supervision mission report, Bank staff attributed the delay in land adjudication to an increasing lack of resources and personnel. While p ocurement for the demarcation work progressed smoothly, the Land Ministry's budget for the field and office equipment needed for geodetic measurement, mapping and legal work had been eliminated as part of a more general Government austerity program. 69. An October 1983 supervision mission noted that MUDP had not released funds to the ML&M Adjudication Division out of fear that project resources and equipment allocated for adjudication purposes would be diverted to other uses. While the adjudication teams -- despite inadequate field staff and management -- did make some progress in Sonewein and Slipway, they increasingly fell behind schedule. Of the 1,700 parcels of land surveyed under the Project, only 241 titles were registered, according to observations made during the June 1986 MUDP National Seminar on Urban Development and Management Strategies. 70. According to the PCR (para. 6), land adjudication activities were hampered by lack of support from the highest officials of the ML&M. Conflict within the Bureau of Lands and Survey in 1986 led to the suspension and eventual dismissal of the original referee. Fear of property taxes may also have been a factor. After 1984, no claims and no registrations were made, even though 228 parcels of land were demarcated in 1985. According to a 1986 paper by the ML&M Research and Training Officer, the people of Monrovia prefer to live with the "old devil of deed registration," which they know, rather than the new Land Registration system, which they distrust. D. Institutional Development 71. Even if the land adjudication component had been more successful, MCC would not have benefited inasmuch as it was never effectively enabled to collect property taxes. In March 1982, Liberia's Head of State signed the decree agreed upon during Negotiations empowering MCC to start collecting city-wide property tax revenues, retaining 30 percent of the funds collected. However, it is likely that this was done essentially because the Bank tied effectiveness of the IDA credit to this condition. Government officials expressed misgivings that this action might be considered a precedent for secondary cities and other municipalities in Liberia. 18 72. Later supervision reports noted that the Government's decree was not implemented and that property tax revenues were never made available to MCC, despite repeated promises that this clear violation of the Credit covenant would be corrected. Consequently, MCC -- along with Liberia's other local governments -- continued to be largely dependent on Central Government grants. Only a few minor fees and taxes (eg. garbage, cemetery and court fees, etc.), amounting to about 4 per cent of total local revenues, remained under the direct control of cities and counties. Particularly disturing to MCC was the transfer of responsibilities for markets from local government to the new Liberia Markets Association (LMA) without clear arrangements for expanding and maintaining the markets or for collecting garbage generated there. 73. Because accounting systems were so weak, the Government had difficulty tracing its funds, a factor which considerably delayed payments to agencies responsible for delivery of local services in Monrovia. In 19856, Government efficiency was reported to have diminished to such an extent that the salaries of officials in a number of agencies had not been paid for several months. Reflective of the generally deteriorating financial situation, counterpart funding allocations received by MUDP declined from US $500,000 in 1983/84, to US $200,000 in 1984/85 to nothing at all during 1985/86. Delays in payment of staff salaries encouraged inaction and absenteeism and impeded the provision of municipal services, such as garbage collection, which, in turn, led to decreased revenues for these services. 74. In addition, municipal officials were intimidated by the possibility of political intervention from high levels or by the fear of being dismissed, imprisoned or harmed. Since many top officials were political appointees drawn from the army, with little previous management experience, they were largely unwilling or unable to focus on urban policy matters. As a result, many City Ordinances having to do with public health, zoning and use of sidewalks and streets were not enforced. In a paper presented at the 1986 MUDP National Seminar On Urban Development and Management Strategies, Liberia's Assistant Director for Environmental Health pointed out that more pollutants and toxic substances were being introduced into the local environment and that overall sanitary conditions in Monrovia were steadily worsening. E. Project Costs and Disbursements 75. In comparing project costs, as presented in the 1982 Development Credit Agreement, with disbursements, as summarized by World Bank Loan Department in September 1988 (Annex 2), it should be noted that SDRs rose in value by more than 11%, as against the U.S. dollar over this period. With this in mind, there appears to be a close correlation between anticipated and actual costs in SDR amounts for the following categories: Civil Works Credit Agreement Disbursement (a) MCC 3,540,000 3,783,103 (b) LWSC 460,000 491,578 19 Consulting Services (a) MCC 1,210,000 1,330,924 (b) LWSC 130,000 144,259 Project Preparation 440,000 441,632 76. Of the SDR 8,900,000 allocated for the project, only SDR 7,109,310 was spent, leaving a balance of SDR 1,790,689. Equipment intended for solid waste management was never purchased because of the lack of interest on the part of the Borrower in this component. Funds intended for consultancy services and training necessary to implement the recommendations of 1984 Price Waterhouse municipal manigement report were never spent because of the opposition of the Government to carrying out these recommendations. 77. Annex 3 compares the disbursement schedule, as presented in the Staff Appraisal Report, with the actual disbursement rate documented by the World Bank Loan Department in September 1988. It should be noted that, by the end of 1984, less than 40% of funds had been disbursed, as against 97% expected to be spent, according to the Appraisal Report. The institutional and financial difficulties contributing to these delays were summarized in the preceding sections. During the period from March 1985 to March 1986, nearly 40% of total disbursements were made. The delay in the release of funds caused the salaries and operating costs of MCC for project-related activities to go up by more than 43%, from SDR 530,000 to SDR 758,484.96. IV. PROJECT RESULTS A. Physical Accomplishments 78. The Audit has no reason to doubt the validity of the project achievements listed in the PCR (para. 9) and mentioned above in the implementation review. In a paper prepared for the 1986 MUDP National Seminar on Urban Development & Management Strategies, the head of the Energy and Utilities Section of the Ministry of Planning and Economic Affairs reported that the Monrovia Urban Development Project had brought direct benefits to more than 130,000 urban poor, or about 30 percent of the residents of Monrovia. Assuming this conclusion is correct, 30% more people were benefitted by the Project than anticipated in the Appraisal Report. This accomplishment is remarkable, considering the economic and political situation existing in Liberia during the period of project implementation. 79. Because this is only a Desk Audit, there is no way to independently evaluate what was accomplished, and the statistical information available in Bank files is incomplete. However, a comparison of the "upgrading components" listed in Annex 1 with the MUDP Construction Completion Report for October and November 1985 indicates that most of what was planned was actually built within the projected budget, when physical and price contingencies caused by unforeseeable implementation delays are taken into account. According to the 1986 Annual Report of MUDP, except for five wetcores, most of the facilities within the communities selected for upgrading were actually completed. In 20 addition, as planned, safe drinking water was piped to other low-income communities through the LWSC standpipe extension program. B. Economic Reevaluation and Financial Performance 80. In the Appraisal Report, the economic rates of return for the three upgrading areas of Claratown, Sonewein and Slipway were estimated, in the aggregate, to be about 12%. This rate was calculated on the basis of anticipated increases in land values due to project-sponsored upgrading. Non- quantifiable benefits included expected decreases in morbidity and mortality and increases in labor productivity due to public health improvements associated with project sanitation works. 81. Because of the failure of the land adjudication component, however, the actual increase in land values due to the project cannot be determined. Although Bank staff noted efforts of residents in the upgraded communities to improve their houses, they did not quantify these benefits. Unfortunately, there is also no quantified evidence of improvements in health levels or labor productivity in the areas affected by the project. Nevertheless, as pointed out above, 130,000 people were reported to have directly benefitted from the project, as against about 100,000 expected to be benefitted in the Appraisal Report. Given the higher number of beneficiaries relative to roughly the same level of costs anticipated at appraisal, the economic rate of return of the project may, in fact, be slightly higher than originally estimated. However, in the absence of more precise figures, the PCR did not attempt to reestimate the rate of return (para. 9) and the Audit has no independent basis for assigning a numerical value to it. 82. While the operation probably did benefit most of the low-income residents of the upgraded communities, it may have adversely affected the financial situation of the institutions responsible for the project. As pointed out in its 1986 Annual Report, MUDP was unable to generate revenue, except for petty funds from services at the wetcores in the project areas. As has been observed above, moreover, MCC was never allowed to collect property taxes or other additional revenues. Because of the controversy (discussed below) over the most effective way of charging for water use in Monrovia, the LWSC also lost money on each gallon of water that it produced. Due to apparently limited cost recovery, it is doubtful that the Liberian Government benefitted financially from the Bank's Monrovia Project. Consequently, the Audit must conclude that the financial benefits of the project remain potential -- rather than actual -- to the institutions undertaking this operation. C. Institutional Performance 83. The MUDP MUDP performed well in planning the steps essential for implementing the Monrovia project, in supervising the work of contractors and in overcoming the many problems that were encountered. Its staff, which consisted of largely capable and interested professionals, also worked well with the three communities selected for upgrading. The MUDP Annual Report for 1986 noted that its Community Development Section was successfully able to strengthen the capabilities of existing community associations. Residents of 21 the project communities were assisted in formulating a "service charge," based on the residents' willingness and ability to pay for services rendered. This was expected to be a first step towards cost recovery. The essentially positive performance of MUDP demonstrates what might be accomplished by local government in Liberia if greater decentralization were permitted by the Central Government. 84. The MCC In July 1984, Price Waterhouse Public Accountants completed a six-month study of the operations of MCC, including an analysis of its statutory laws, organizational structure and management, revenues, accounting systems and staff development programs. The six volumes -- which were presented in preliminary form to the MUDP steering committee in October 1983 -- required 21 consultancy staff months and cost nearly US $300,000. The most important of the many recommendations contained in the Executive Summary of the study included the need to: (1) transfer many responsibilities from Central Government to MCC; (2) minimize the overlap between Central Government and MCC functions and services; (3) restrict the City Police Force and the City Court to responsibility for the enforcement of city ordinances; (4) incorporate an elected Mayor and Council; (5) establish seven functional departments and three semi-autonomous departments; (6) increase MCC9s share of property revenues to 100 percent; and (7) have ACC assume full control over market management and the marketing associations. 85. According to Bank supervision missions, no follow-up actions were ever taken on the recommendations of the Municipal Management Study. It did lead, however, to the first "National Seminar on Urban Development and Management Strategies" in June 1986, sponsored by MUDP/MCC, in collaboration with USAID, Price Waterhouse, the UNDP and the various Government ministries and agencies concerned with urban affairs. While there were no funds for Bank participation, representatives from all the ministries and agencies involved in the Monrovia project attended the three-day seminar, along with representatives from major Monrovia corporations, civic groups and other international organizations. The weaknesses of MCC were clearly pointed out, including the following: (1) of MCC's 700 MCC employees, only four were college graduates; (2) the appointed Mayor and Council were less responsive to the electorate than to those appointing them; (3) the City's various development programs were largely uncoordinated; (4) building codes and municipal ordinances were seldom properly enforced; (5) MCC continued to be denied the right to collect property taxes; and (6) strict fiscal discipline at the local level was lacking. 86. Detailed training recommendations for all levels of municipal management also emerged from the 1984 Price Waterhouse report. In addition to a combination of seminars, courses and workshops for various levels of staff, the study emphasized the importance of linking course attendance to a merit system, requiring job competence to be demonstrated for status, salary and position. Altogether 133 man-months of training were recommended over a six year period -- including 80 months of overseas training and 45 months of local training -- costing about US $265,000. 87. While these training recommendations went largely unimplemented with the suspension of all Bank disbursements to Liberia in 1986, the Monrovia 22 Urban Project Director continued to request Bank assistance for various short- term training programs. Bank staff recommended an exemption from the ban on disbursements in order to continue the technical assistance, but -- because of the misuse of funds advanced to Special Accounts held with the National Bank of Liberia -- the Bank declined to authorize further replenishments. MUDP staff, however, did continue to attend a number of training courses sponsored by European foreign aid programs. 88. The LWSC Under the Monrovia Urban Development Project, the Barlatt tariff study for water supply was carried out in 1984. The study recommended a surcharge on property taxes, combined with metering of selected consumers to monitor actual consumption. Because of the weakness of the property tax system in Monrovia, this recommendation was found unacceptable to Bank staff. Thus, the most effective way of charging for water use in Monrovia remained undefined. Consequently, many of LWSC's standpipes were either disconnected for non-payment or illegally connected. According to the 1985 Bank Appraisal Report for the Monrovia Second Water Supply Project, LWSC's financial situation worsened with each gallon of water produced and distributed. The billing and collection system was undermined by erroneous entries, uncollectible accounts, mismanagement of personnel and unwillingness of Government agencies to pay for services. 89. The Second Water Supply Project (Cr. 1563-LBR) provides a follow-on to the Monrovia Urban Development Project, as well as to the First Water Supply Project (completed in 1983). Based on the Bank's experience with earlier projects in Liberia, the Second Water Supply Project requires that a substantial amount (US $550,000) of the counterpart funds be generated by LWSC and deposited in a project account before major construction components of the project are undertaken. It also provides funds for repairing and securing standpipes (through locked kiosks) to support privatization. The last Bank supervision mission (October 1987) indicated that the LWSC continued to have serious managerial problems. However, with the financial assistance of the West German Government (GTZ), Bank staff hope that the project can be completed in 1989, despite the termination of Bank disbursements in July 1987. D. Project Sustainability 90. The Audit notes the skepticism of Bank staff regarding the sustainability of what was accomplished in Monrovia. Without improved financial control and resource mobilization, the physical investments undertaken by the project (the least important of the project objectives) are unlikely to be adequately maintained. According to the final Bank supervision mission (April 1986), MCC continued to be characterized, not only by an unreliable and inadequate revenue base, but also by unsatisfactory financial, planning, budgeting and accounting systems. As a result, it lacked control over expenditures, causing MUDP to have trouble getting payments on their vouchers, as well as for staff salaries. The refusal of the Central Government to turn over responsibility for the collection of property taxes-- as provided for in the Credit Covenant -- had a demoralizing effect on MCC, as well as arousing public resistance to payment. While the property tax comprises only 1% of total tax revenues, the fact that it is poorly 23 administered by the Ministry of Finance severely reduces its potential. E. Bank Performance 91. In assessing Bank performance, the Audit must reiterate three facts: (1) the successful completion of most of the planned civil works within the anticipated budget; (2) the difficulty of quantifying the benefits derived from the project; and (3) the unlikelihood that the project is sustainable. The work of Bank staff in both the preappraisal and supervision stages was commendable in facilitating the completion of the physical works within the planned budget, despite the difficulties encountered. With regard to the other components, the following extenuating circumstances should be noted: (1) the pressure on Bank staff to get the project approved as quickly as possible; (2) the tendency to minimize political and economic risks and to underestimate the import6nce of political realities; (3) the Bank practice of using a "blueprint," rather than a more incremental and experimental approach; (4) the lack of funds for adequate project supervision and evaluation; and (5) the need to terminate this project and relations with Liberia before follow-up projects could be developed. 92. A number of useful studies were undertaken by consultants in preparing this project, including analyses of the housing component (May 1979), informal sector employment (June 1979), traffic engineering (December 1979) and urban poverty. Bank staff, nevertheless, underestimated the importance of undertaking a more comprehensive political risk analysis. While the political hardware (constitutional structure of Monrovia) was theoretically in place, the necessary political software (underlying political relationships) was largely ignored. Unless MCC gained real political credibility, it was unlikely that people would be willing to pay taxes to it. Yet, the necessary political preconditions for obtaining the legitimacy of Monrovia's local government -- particularly the carrying out of mayoral elections -- remained unemphasized. 93. The experience with this project also suggests the occurrence of lost opportunities to mobilize public resources and tap the community's desire for better living conditions. Many households reportedly contributed cash to local chiefs to pay for neighborhood improvements such as road and footpath extensions, drainage, health and day-care services. Community committees were reported, in 1986, to be instrumental in developing the concept and method for "service charges." The Claratown Development Association was particularly successful in raising funds for such activities as development control, maintenance of roads and drains, sanitation and solid waste disposal. 94. The use of private contractors to provide urban services might also have been a possibility. USAID, for example, successfully used a reliable and competent Liberian contractor to operate the public toilet facilities, a small market and a small rental housing project in the West Point area, fully covering costs by charging for services and collecting rents and fees. The Bank also experimented with this approach under the Second Water Supply Project with the privatization of standpipes in Logan lown. This approach worked well until November 1985 when the kiosks were vandalized in the aftermath of a coup attempt. However, the potential usefulness of this 24 approach also suggests, in retrospect, the need for closer coordination between the Bank's urban development and water supply specialists. V. CONCLUSIONS AND LESSONS LEARNED A. Conclusions 95. Because the physical achievements of this project are unlikely to be sustainable, and inasmuch as the land adjudication and institutional components clearly failed to meet their objectives, the project as a whole c&nnot be considered a success. Nevertheless, within the parameters of prevailing Bank procedures and practices in the early 1980s, the Liberian urban project was innovative, the proper preparatory steps -- including the use of consultants and the funding of feasibility studies -- were followed and a successful Project Launch was undertaken. Bank staff established a well- functioning project unit and corrected problems that threatened to undermine implementation of physical works that cubsequently were successfully completed. Yet, inasmuch as the project failed to achieve its financial and institutional objectives, questions must be raised as to what went wrong. In drawing its conclusions, the Audit associates itself with the following observations -- as revealed in personal interviews and the project files-- made by some of the Bank staff involved in project supervision: a. The timing of components Because the civil works -- the aspect of greatest interest to the Government -- were largely completed by 1984, the financial and institutional reforms specified in the Credit Agreement and recommended by the 1984 Price Waterhouse study were largely ignored. Furthermore, the Bank lost all leverage with the Liberian government once the civil works were completed. This prevented it from being able to apply more effective pressure on the Borrower to comply with that part of the Credit Covenant which required that MCC be allowed to collect property taxes within the city. If the civil works could have been delayed, pending institutional reforms, the project as a whole might have been considerably more successful. b. The integration of components Ideally, the investments and approach followed in the Second Water Supply Project for Liberia (appraised in February 1985) should have been introduced at the time of the Monrovia Urban Development Project. The Second Water Supply Project required LWSC to improve its billing and collection system, introduce a standpipe privatization plan, institute an affordable charge structure and ensure that standpipes were well maintained so that a maximum number of users could be served. Until these reforms are institutionalized, moreover, it is likely that the improved urban services introduced under the Monrovia Urban Development Project cannot be sustained. By integrating components, the Bank might also have facilitated greater cooperation among Bank specialists and divisions undertaking complementary projects. 25 c. The avoiding of politics It is Bank policy to avoid intruding into domestic political matters. The growing Bank emphasis on institutional reform and project sustainability, however, frequently forces Bank staff into an untenable position when they cannot insist with national leaders on the necessary political pretonditions for reform, including legitimacy, creditability, two- way flows of communication, a meritocracy, protected spheres of authority, etc. In this context, moreover, it is an illusion to expect people to pay property taxes or service fees if enforcement is weak, service provision is poor, collection is mismanaged and funds are wasted. While the dangers of alienating Borrowers must be recognized, more open discussions about the preconditions for 3kstitutional and financial reform should not be avoided even when these are politically sensitive. d. The need for flexibility It is impossible to determine in advance the opportunities and difficulties that will arise in the course of project execution. Different approaches will work to different degrees at different times depending on local circumstances. Bank staff involved in supervising this project felt the need to be able to experiment with alternatives and to respond quickly and appropriately to project-related opportunities. As indicated above, existing community associations and local NG0s were able and willing to cooperate with the Monrovia project. A much greater effort could have been made to work with these groups, taking advantage of their contacts with the public and their knowledge and experience. Flexibility is also needed in selecting project components for which people are willing to pay. If Bank staff could have taken advantage of these alternatives -- rewarding with funds those possibilities that offered the greatest potential of being self-sustaining -- the Monrovia project might have been more successful. e. The importance of motivation Richard Heaver and Arturo Israel in their 1986 Bank study entitled "Country Commitment to Development Projects" emphasize the importance of building commitment, using incentives, isolating opposition, associating progress with performance, building on a series of small operations with future assistance dependent on results and releasing loan tranches based on progress made. In building commitment, the Bank needs to maintain a high level of incentives -- primarily in the form of funds for projects considered important by Borrowers -- in order to more successfully bargain with them for the attainment of needed institutional reforms and other policy objectives. In the Monzovia project, the loan amount was insufficient to motivate such reform or even to generate serious governmental discussions and consideration. Nor was it apparently large enough to arouse widespread public enthusiasm for the project. Because the loan amount was small by Bank standards, moreover, the Bank itself appears not to have been sufficiently motivated to spend greater time on the supervision of project implementaticn. 26 E. The inadequacy of supervision Bank staff felt that their periodic visits to Monrovia were not long enough t3 provide adequate supervision for the project. There was a clear reed for an in- country officer responsible for monitoring and supervising implementation, for facilitating communication between the public and project staff, for harmonizing the Bank's urban sector projects and for coordinating with other foreign donors active in local urban development. A resident Bank manager, moreover, might have facilitated some of the participant-observation monitoring and evaluation found useful in similar Bank urban projects elsewhere. B. Lessons Learned 96. In the PCR for this project, the Region affirms (para. 15) that "project objectives should be scaled to what is more realistically possible in the absence of major macro-economic reforms." While the Audit accepts this conclusion, it would also liked to emphasize the need to encouiage operational procedures conducive to the institutional reforms essential for the successful achievement of project objectives. More specifically, the lessons emerging from this project with regard to successful implementation and institutional development are: a. The importance of step-by-step procedures The results of the Monrovia project confirm the validity of the incremental approach suggested in the 1987 OED report concerning the Bank's experiences with rural development. Under this approach, each phase of a project could be carefully marked out, with an understanding that each successive step would be taken only after significant progress had been achieved with regard to the previous one. Such an approach was, in essence, adopted in the Second Water Supply Project for Liberia, where the execution of physical components was deferred until certain institutional and financial targets had been met. Had this approach been followed in the Monrovia Urban Development Project, Step 1 could have focussed on institutional development-- particularly the establishment of cost-recovery arrangements-- essential for the success of Step 2, which could have concentrated upon infrastructural investments and community facilities. Step 3 could then have included a variety of other components (traffic management, education, small-scale enterprises, etc.) that were left out of the project. The entire project, accordingly, could have been presented in the form of a larger multi-sectoral urban loan package, with funds to be released in tranches conditional upon attaining sufficient progress in institutional development. During or after each step, the project could have been evaluated and, if necessary, reconsidered. This would allow for contingencies not 27 anticipated at Appraisal while, simultaneously introducing greater flexibility into project implementation. 8/ b. The need to take advantage of urban opportunities The willingness of peop-e living in Monrovia's low and middle income settlements to form Community Health Motivation Councils to improve drainage, sanitation, and solid waste management systems was pointed out in a 1979 consultant's report. The potential of community groups to undertake neighborhood improvements could have been facilitated, according to a 1985 Bank supervision mission, through establishment of a community developaent loan fund and/or a matching grant system for local infrastructure and services. The existing MUDP unit might have managed this fund, evaluating proposals and assisting the communities in carrying them out -- perhaps for a fee -- with the power to cut off services to uncooperative localities. Another unexplored possibility would have been to use Liberian contractors to manage public facilities and cost recovery arrangements, as was, in fact, done by USAID in Monrovia's West Point settlement and by the Bank in the Second Water Supply Project. c. The need for greater openness with the Borrower Liberia's institutional barriers to urban development -- and particularly to administrative and financial decentralization for the provision of urban services -- should have been clearly discussed with the Borrower. These concerns include such persisting characteristics as excessive patrimonialism, formalism, centralization and inefficiency in the use of manpower and other resources. As it was, the crucial linkages between institutional reform, cost-recovery and the sustainability of the flow of socio-economic benefits resulting from project-financed physical improvements were never clearly understood or seen as essential by the Borrower. d. The need for more adequate risk assessment While much useful research was done at the preappraisal stage, these studies failed to revepl -- or properly assess -- many of the political, institutional and financial difficulties encountered during project implementation. If the Bank had more fully taken these problems in account at the Appraisal stage, in all likelihood, it would also have defined better procedures to minimize and overcome risks. 8/ One Bank Division, in revieving a draft of this report, observed that the proposed method of enhancing the Bank's leverage through step-by-step procedures might have been "unrealistic" in this particular case given the complexity of urban management in Liberia which directly or indirectly involves a "large number of central government ministries, departments and agencies and weak local goverments." The same reader further noted that "some project components -- such as the land adjudication component in Monrovia-- concern the administration of fundamental constitutional rights," while others, like property tax reform, "involve changes in the basic fiscal structure of the country." 28 e. The need for more intensive supervision To have been more effective, the periodic supervision missions required the assistance of a resident manager. Such a manager could have been responsible for all the Bank's urban and water projects in Liberia as well as for liaison with other donors involved in urban development. With selection of the right sort of person-- concerned with community, institutional and financial development as well as the physical aspects of project implementation --the Bank might have been in a better position to encourage achievement of the multiple objectives sought in this operation. In any event, in undertaking risky projects such as this one, the Bank will have to spend greater effort and resources on supervision if it hopes to substantially improve their chances of success. 29 Annex 1 T4tal Priftet Co'te Jamaa 1981 vocal oarelgs Total "te UTcRArDIc * Claaus a-Sls Ifrantruccure 466 547 1,033 55 Off-Sie lfåragtructure 35 79 114 70 Primary 'chool 163 38 201 19 wealth h-tt 19 5 24 20 Come~y Ceer 26 5 31 15 KMMLkh 211 58 14§ 22 Sa&susta: Cm Sita Infrastructure 504 516 1,020 50 Off sItu wtfrascTucture 154 350 504 70 ,raary 5chol laprovesmta 54 36 90 40 arltheCtt la 6 24 25 ORET centar 26 5 31 15 Mar.t Zzpa 140 42 182 23 Sch-ToaL Iw 3- 151 -5 Slipusy: <b site Infratructure 340 296 636 47 Off Sita Iafrastruetre 21 46 67 70 Prary School Inprovmete 26 16 42 40 esalth vue 18 6 24 25 Cffim y Center 27 4 31 15 -Total -ER 44 Other& ~hustaa an 93 23 116 2 mealth C:ar Nu Krl t~. 377 200 577 35 &umag ~ 4ipmet & Ca~ ~ - 50 5i0 100 1uC z~andpipe Ecteneions 182 418 600 70 rtal T~etography a apinc • 12C 120 100 D saäris b operattas 68 143 31 17 Su-Totua144l.3 sunr .am a. iL 1~ 3-588 7.147 5 LAM ~1~DC6IM ftae I: Proje Ara 30 121 427 PMas M: City utde _66, a56 1.122 12AL 972 3 1.549 37 Tratas 3 127 127 im gea 57 514 571 90 Tedna~ AuIstauce 26 232 258 90 9~ a~y&'s • 450 450 im -.m -.. - .-. UB-TAL- (1ASE) 4,643 5,459 10,102 54 4~hy~ical Coa. 343 3$7 700 51 Tr;.e cet. ifi -4U 255 rAL. PVMT cost 6.71 6,761 1237 1 Source: Staff Appraisal Report, November 30, 1981. * Total Upgrading Costs: US$4,323. 30 Annex 2 CREDIT CATEGORIES Allocation Disbursed Balance Category (SDR Equivalent) /a SDR SDR b 1. Civil Works (a) MCC 3,540,000 3,783,103 6,896 (b) LWSC 460,000 491,578 421 2. Goods/Services 90,000 3. Equipment (a) Borrower 580,000 8,805 571,195 (b) MCC 220,000 85,861 134,139 4. Salaries/Costs (a) Borrower 710,000 8,128 101,872 (b) MCC 530,000 758,485 1,515 5. Consulting Services/Training (a) Borrower 520,000 8,196 352,803 (b) MCC 1,210,000 1,330,924 379,076 (c) LWSC 130,000 144,260 0,00 6. Miscellaneous 910,000 441,632 283,110 Total 8,900,000 7,109,310 1,790,690 /a US Dollar per SDR: April 1982, 1.12967. /b US Dollar per SDR: September, 1988, 1.27964. (11.3% Rise in Value of the SDR over the US $ during the project). Sources: Development Credit Agreement, April 21, 198k; Bank Loan Dept., September 13, 1988 31 Annex 3 ACCUMULATED DISBURSEMENT SCHEDULE Period Date Anticipated % Actual % 12131/82 37 9.6 06/30/83 62 12.1 12/31/83 79 15.5 06/30/84 91 29.0 12/31/84 97 39.9 06/30/85 100 54.9 12/31/85 68.6 06/30/86 79.1 12/31/86 79.7 06/30/87 79.9 Sources: Staff Appraisal Report, November 30, 1981; Bank Loan Department, September 13, 1988 - 32 - PROJECT COMPLETION REPORT LIBERIA MONROVIA UPBAN DEVELOPMENT PROJECT (Credit 1223-LBR) IR II II Africa Regio - 33 - LIBERIA MONROVIA URBAN DEVELOPMENT PROJECT CR. 1223-LBR PROJECT COMPLETION REPORT Background 1. The Monrovia Urban Development Project, conceived as a result of a study (the Monrovia Urban Study) carried out in 1976-78, was intended to complement the efforts to improve urban infrastructure and utilities, particularly those serving the poor, of other ongoing projects it the water supply and power sectors. Appraised in 1979. the project was delayed and modified because of the military coup on April 12, 1980. Originally projected to begin about mid-1981 and be completed by June 1985, the project was not declared effective until October 21, 1982 and the closing date was extended twice, first to June 30. 1986, then to December 31, 1987. Proiect Objectives 2. The project had three major objectives: (i) improving living conditions of the urban poor; (ii) strengthening municipal management, including development of a delivery system for upgrading at the municipal level, and improving the city's revenue base; and (iii) strengthening property tax collections requiring establishment of a land ownership registry in Monrovia. Project Description 3. The project comprised three major components: Upgrading infrastructure, utilities and social service facilities in three high-density, low income areas suffering acute deficiencies (Slipway, Sonewein and Claratown) and extending water distribution to unserviced areas (e.g. New Kru Town, Logantown, Jamaica Road, etc.); institutional development aimed at strengthening Monrovia City Corporation, particularly its financial, administrative and technical services management capacity; and a land adjudication component to complete a land title registry to enhance the land market and property tax administration. - 34 - Project Implementation 4. After the establishment of a relatively well qualified and dedicated department, the Monrovia Urban Development Project (MUDP) Department, as an autonomous self-accounting unit within the Monrovia City Corporation (MCC), project effectiveness was postponed because of "elays in obtaining executive approval of a decree transferring a portion (30Z) of property tax receipts to MCC. Final detailed designs and contract documents were also delayed. Design standards which had been agreed upon at the time of appraisal were scrutinized and subjected to further review by the Ministry of Public Works. As a result, the initial designs had to be modified and tendering took place one year later than had been estimated at appraisal. Despite initial delays, chronic shortages of counterpart funds, political interference and turmoil, the project's civil works and community facilities with only minor exceptions were completed within a reasonable time and within budget. 5. From the very beginning the project was affected by the Government's deteriorating financial situation. Counterpart funds were never made available to the project unit in a timely manner and actual receipts were between one half and two thirds of budgeted requirements. Transfers of property tax receipts, originally intended to finance capital investments and be used as counterpart funds, were diverted by MCC for other uses. MCC's financial base was further eroded by the transfer of market revenues, which had been a major source of finance up to 1980, to the Liberia Market Association (LMA). The Ministry of Cooperatives and Rural Development, in fact, built a number of new markets and extended the Rally Time Market at great cost to the public purse, before handing over these facilities to the LMA without any provision for recovering costs. I 6. Collections of property taxes declined, in part due to the generally declining economic environment, but also because of the exodus of many landlords following the 1980 coup and the lack of incentives for the Ministry of Finance to improve collections rates, part of which would have accrued to MCC. The land adjudication component, which had been intended to stimulate regularization of tenure, and hence indirectly to enhance potential property tax receipts, did not have the full support of the highest officials in the Ministry of Lands and Mines, itself beset by internal problems, and fell behind schedule. 7. At the same time, and throughout the period of implementation MCC lacked the management, staff and organizational capacity to see beyond the day-to-day crises and deal with strategic issues. Lacking an elected and therefore representative council and Mayor, MCC also lacked grass-root political support and credibility. The project financed an urban management study, carried out by Price Waterhouse Ltd., to recommend improved organization, administration and financial management of MCC, including necessary legal reforms, training and staffing requirements. The study's recommendations, some of which entailed public debate of policy issues, while others were concerned with strictly operational matters - 35 - within MCC's mandate and responsibilities, were not acted upon until the end of the project. 8. At all stages of project implementation, but particularly at times of tendering for major civil works contracts, MUDP's Director and senior staff were under considerable political pressure to bend the Bank's procurement procedures. It is to their credit and professional integrity that civil works were generally completed within budget and almost on schedule. Achievements -- and Failures 9. The residents of New Kru Town have now access to safe drinking water; Claratown is accessible to public transport, has roads, water, sewerage, school, health facilities, and communal toilet-bathroom blocks. Similar, improvements have been made in Slipway and Sonewein. This is testimony to the physical achievements of the project. However, there are insufficient data available to allow recalculation of the rate of return. 10. The deteriorated state of MCC finances, lack of policy and organizational reforms and lack of improvements in MCC's technical and operational capacity to provide services and maintain infrastructure exemplify the project's failure in the non-physical realm of *institutional development'. It is worth noting, however, that even on the institutional front the project proved to be the seminal catalyst for progress. The MUDP Department, while drawing much of the talent from what was left of HCC, developed into a strong, motivated body of staff that gained considerable experience in project planning and administration covering community development, engineering, architecture and accounting. 11. At the grass-roots level, the Claratown Development Association has, as a result of the project and because of the weak MCC administration, taken over a number of municipal functions such as development control, maintenance of roads and drains, sanitation and solid waste management. It has done this using a cooperative, community-based approach. In fact, it has even mobilized residents' financial contributions for capital improvements. 12. Concerning the third major objective, the resolution of the longstanding problems with land adjudication proceeded slowly, but little actual progress was made in the area of property tax collection. Although the necesssary decree to give MCC authority to collect property tax had been passed early on in the implementation phase of the project, it was never acted upon due to political and administrative resistance. Lessons Learned 13. The project was implemented at a time of political turmoil and a chronic worsening financial crisis. In this environment the interest and ability of government to focus on urban policy matters was low. The project was short of counterpart funds, and when the country could not meet - 36 - its debt obligations, all Bank disbursements were suspended. This happened twice during the course of implementation and resulted in significant additional costs of those contracts affected. 14. The day-to-day operations of government were regularly being short-circuited by political intervention from highest levels and many officials were unwilling or unable to make decisions for fear of recrimination or worse. Responsibilities and accountability between ministries and between different levels of government were ill-defined, and ultimately highly centralized. 15. In hindsight it is obvious that institutional responsibilities and procedures should be clearly delineated at the outset and constantly reviewed. Macro-economic issues should be recognized and project objectives should be scaled to what is more realistically possible in the absence of major macro-economic reforms. 16. Institutional development - in its broadest sense meaning the improved administration and management of human, material, and technological resources - is a long term process which should be seen as an ongoing activity. The key elements of this component - technical assistance and training - should be initiated as early as possible in the project cycle. The definition of this component should ensure the preservation of its outcomes, developing in the institutions an improved ability to manage and *make things work* in the future, beyond the life of the project.