Document of The World Bank FOR OFFICIAL USE ONLY Report No: 22446 IMPLEMENTATION COMPLETION REPORT (IDA-24790) ONA CREDIT IN THE AMOUNT OF US$20 MILLION TO THE MOZAMBIQUE FOR A RURAL REHABILITATION PROJECT June 12, 2001 Rural Development Operations Eastern and Southern Africa Africa Region |This document has a restricted-distribution and may be used-by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective ) Currency Unit = Mozambique Metical At appraisal Mt. 1,450 = US$ 1 US$ 1 = Mt. 16,600 At Completion FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS ARDP Agricultural Rehabilitation and Development Project ASRDP Agricultural Services Rehabilitation and Development Project BPD People's Development Bank DINAGECA National Directorate for Geography and Cadaster DNER National Directorate of Rural Extension FM/Fl Farmer-Managed/Farmer-Implemented FSR Farning Systems Research INIA National Institute for Agricultural Research IPM Integrated Pest Management JC Joint-Venture Company M&E Monitoring and Evaluation MADER Ministry of Agriculture and Rural Development PROAGRI Agricultural Sector Public Expenditure Program QAG Quality Assurance Group RRP Rural Rehabilitation Project T&V Training and Visit Extension Vice President: Callisto Madavo Country Manager/Director: Darius Mans Sector Manager/Director: Sushma Ganguly Task Team Leader/Task Manager: Miriarn Brandao/David Nielson FOR OFFICIAL USE ONLY MOZAMBIQUE RURAL REHABILITATION CONTENTS Page No. I. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 4 5. Major Factors Affecting Implementation and Outcome 10 6. Sustainability 12 7. Bank and Borrower Performance 13 8. Lessons Learned 14 9. Partner Comments 16 Annexes Annex 1. Key Performance Indicators/Log Frame Matrix 17 Annex 2. Project Costs and Financing 19 Annex 3. Economic Costs and Benefits 21 Annex 4. Bank Inputs 22 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 24 Annex 6. Ratings of Bank and Borrower Perforrnance 24 Annex 7. List of Supporting Documents 25 Annex 8. Government of Mozambique Implementation Completion Report 30 MAP (S) Map of Mozambique 43 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization. 1. PROJECT DATA Name. RURAL REHABILITATION L/C/TF IDA-24790 Number: Country/Department: MOZAMBIQUE Region: Africa Regional Office Sector/subsector. AG - Agency Reform; VM - Natural Resources Management; WR - Rural Water Supply & Sanitation KEY DATES Original Revised/Actual PCD. 11/16/90 Effective: 09/01/93 12/08/93 Appraisal: 06/21/92 MTR: Approval: 03/30/93 Closing: 12/31/98 12/31/2000 Borrower/Implementing 4gency. Government of Mozanibique (GoM)/National Institute for Rural Development (INDER) originally - and then, the National Directorate for Rural Development (DNDR) under the Ministry of Agriculture and Rural Development (MADER) Other Partners: National Directorate for Water (DNA) and National Directorate for Geography and Cadaster (DINAGECA) STAFF Current At Appraisal Vice President: Callisto Madavo E. Jaycox Country Manager: Darius Mans S. Denning Sector Manager: Karen Brooks C. Helman Team Leader at ICR. David Nielson D. Cotlear ICR Primary Author: FAO 2. PRINCIPAL PERFORMANCE RATINGS (HS=Highly Satisfactory. S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: M Bank Performance: S Borrowver Performance: S QAG (if available) ICR Quality at Entry: U Project at Risk at Any Time: No 3. ASSESSMENT OF DEVELOPMENT OBJECTIVES, DESIGN AND QUALITY AT ENTRY 3.1 Original Objectives: According to the Staff Appraisal Report (SAR), the primary objective of the project was "to undertake, on a pilot basis, activities to support decentralized economic recovery while creating the capacity and procedures necessary to address the broader post-war situation." The specific objectives of the project were to: (i) build the capacity of institutions responsible for post-war reconstruction and develop procedures for decentralized rural development; (ii) provide land use information and develop land policies for smallholders; (iii) help households displaced due to war to restart agricultural production; and (iv) improve health and life of rural households by increasing access to drinking water. The Rural Rehabilitation Project (RRP) was intended to support Government's National Reconstruction Program (NRP) following a conflict which lasted over a decade and resulted in the massive displacement of rural populations, the destruction of rural infrastructure, the collapse of the economy, and the disappearance of most government services in rural areas. The project was part of a massive donor backing for the Government's rehabilitation effort. However, the weight attached to decentralization and to piloting procedures was premature given the urgent and constantly changing demands of the post-war situation, and the low institutional capacity which existed at all levels of government. The specific project objectives (and related components) were also an uneasy mix of rather disparate interventions for both emergency assistance and longer-term policy development. 3.2 Revised Objectives: The primary project objective was not revised during implementation, but specific project objectives were modified to reflect a 1997 restructuring of the project which eliminated the Distribution of Agricultural Packages component and limited the scope of both the land and water components (see below). 3.3 Original Components: At its inception, the RRP had four main components, namely: (i) Support for Decentralization, including: (i) pilot activities in Sofala and Zambezia Provinces to test procedures for identifying, funding and implementing community- based micro-projects; and (ii) institution-building and training of Government staff, NGOs and community groups; (US$6.4 million base cost). (ii) Land, including: (i) a land study and pilot demarcation for smallholder lands to improve access and secure tenure rights; and (ii) land use mapping and institutional strengthening; (US$2.9 million) (iii) Distribution of Agricultural Packages (Agpacks) to provide seeds and hand tools to smallholders returning to their traditional land after the war; (US$4.5 million). (iv) Rural Water Supply to: (i) increase the coverage of the rural water supply in Sofala and Zambezia Provinces; (ii) strengthen provincial water supply institutions; and 2 (iii) stimulate community participation in operation and maintenance of the systems. (US$5.0 million) The project was to be managed by the then-recently established (1990) National Institute for Rural Development (INDER) which reported directly to the Council of Ministers and held cross- sectoral responsibilities. INDER established a small, contracted project unit in Maputo and two implementation teams led by Provincial Project Advisers (one in each of the two provinces - Sofala and Zambezia). These units were budgeted and financed under the project's decentralization component. Some 190 person-months of long- and short-term international technical assistance (TA) and 150 person-months of national TA were planned to assist INDER in project management and operations, as well as for training, capacity-building and studies. Implementation of the land and water components was to be contracted out to DINAGE,CA and PRONAR respectively. Although the project was designed as a three-year pilot operation, the SAR already anticipated that five years would be needed to complete disbursement. The RRP design rightly allowed for flexibility, in particular with respect to the selection of priority districts within the two selected provinces (depending on the security situation), and the types of community-generated investment ideas (micro- projects). 3.4 Revised Components. In mid 1997, the project was restructured as part of a broader IDA portfolio restructuring, in order to: (i) remove activity overlap between RRP, the Agricultural Services Rehabilitation and Development Project (ASRDP) and the Agricultural Rehabilitation and Development Project, (ARDP); (ii) simplify RRP components; (iii) adjust to the prevailing circumstances (the country was no longer in a post-war emergency situation); and (iv) clear the way for what was to become the agricultural sector development program - PROAGRI. The following main changes were agreed upon: e transfer of the Land Study sub-component to ASRDP; e cancellation of the Agpacks component; s changes in the field implementation arrangements of the Rural Water Supply component to allow private sector participation and Provincial Water Supply Workshops (EPARs) to operate on a ccmmercial basis; and * expansion of the Rural Water Supply component to allow for construction of village- level small piped systems. Total costs of the project were not amended, and the freed resources were re-directed to the Decentralization (micro-projects) and the Land Use Mapping components. The project area for the Decentralization and Water components was extended from partial coverage to coverage of all districts of Sofala and Zambezia. To a great extent, the restructuring agreement sanctioned changes in direction of the project which had already occurred. It simplified the project's structure, making the micro-projects the core focus of the project and other components peripheral. During this restructuring, ARDP Credit was cancelled (See separate ICR reports for ASRDP and ARDP). 3 3.5 Quality at Entry: Throughout preparation and appraisal, there was a difference in emphasis between Government's.objectives for national reconstruction - oriented primarily to restoring infrastructure and agricultural production over as wide a part of the 40 priority districts as possible - and the Bank's desire to pilot systems for community-led rural development. The SAR did not make a clear case for the predominance of the pilot decentralization objective in a project labelled "rehabilitation", and the project, as negotiated, was ambiguous in relation to what the decentralization micro-projects should achieve: whether provision of merely a trial and training ground or a significant contribution to the Government's national reconstruction program. Given the prevailing country situation, the large amount of the credit (US$20 million) was inconsistent with a pilot project and the SAR disbursement schedule (more than 90% over the three first years) was unrealistic. The component coverage of the project was also too broad and complex. The Bank's Quality Assurance Group (QAG) did not review this project. Based on the above, the ICR mission rates the Quality at Entry as unsatisfactory. 4. ACHIEVEMENT OF OBJECTIVES AND OUTPUTS 4.1 Outcome/Achievement of Objectives: The objective of the project was to assist the Government with post-war national reconstruction and to engage with rural communities in improving their livelihoods and living conditions. The impact of the project in terms of national reconstruction has exceeded by far what could have been envisaged from the proposals contained in the SAR. It lasted 7 years instead of the planned 3-5 years, and in its second phase, from 1997, it has contributed not merely to rehabilitation but to rural development. Many of the specific objectives included in the SAR were superseded by the restructuring of the project. Overall project outcome is thus rated satisfactory. The project's most important impact was through the micro-projects. Although original plans called for micro-projects in 8 districts, the project in fact supported micro-projects contributing to improved living standards and education in 27 districts (as summarised in the Table below).2 ZAMBEZIA SOFALA TOTAL Costs Costs |Number ofDistricts 15 12 27 US$ million) (percentage) Schools 88 36 124 5.9 53.6 Health Centers 10 46 56 3.5 31.8 Other Water Points 24 26 50 |} Roads/Bridges 30 7 37 } 0.8 7.3 Markets 12 5 17 0.1 0.9 Livestock 1 14 15 } Agriculture 1 22 23 } 0.2 1.8 Others 18 7 25 0.5 4.5 Total No. Projects 184 163 347 11.0 f 100.0 2 The ICR mission has chosen to cluster micro-projects' contracts which, combined, provide a given service to the community, rather than counting the number of contracts, as was generally done during project implementation. For example, the provision of a school with associated teacher's house, latrines and pump in the same location are treated as a single project; provision of 4 water points to a single community has also been treated as a single project. Source: Estimates constructed by the ICR Mission. 4 Through micro-projects, the construction of school facilities through the project is estimated to have increased the capacity of the rural school system by 50,000 students in a geographic area of some 3,500 km2 It provided health center services (first aid, mother and child care, maternity wards) to a catchment area of around 1 1,000 km2 with a total population of almost 300,0003. The project's contributions are consistent with national and provincial governments' strategies for poverty reduction which give particular attention to: improving access to education for females (which has been shown to be directly related to family welfare); and promoting child health while reducing infant mortality. In addition to their impact in health and education, the civil works contracts financed through the project have generated employment and boosted economic activity in the targeted rural communities and have also provided links with life outside. In some remoter communities, RRP was the first governmental activity to appear after the end of the war to engage with the local populations ancl to invest in improved facilities. The impact of this concrete proof of government goodwill in previously RENAMO-held areas is of inestimable political and social significance, and can only have been positive in the process of nation-building. Provision of social services, opportunities for improved access, marketing and grain-milling must also have promoted the resettlement process. It is interesting that RRP was often perceived as an NGO project, perhaps because of its fieldwork practices with local communities and operational flexibility. The project has also had a significant and positive impact on local capacity to develop and maintain rural infrastructure and skills. Selected local people have been trained, and supported with equipment, in a variety of highly useful skills such as midwifery, carpentry, tailoring, bee- keeping, as well as in those relating to the construction and maintenance of the buildings erected. It has had much less direct impact on capacity-building for decentralization and instituting community-led development (except for rural water supply). For its duration, RRP provided a channel for funds for locally identified investments and it pioneered and created some government capacity for community consultation and organizing community participation. However, the project failed to institutionalise and mainstream the procedures and capacity to undertake decentralized rural development where collective action through local government is required. However, independently from RRP, the Government pursued a pilot for decentralized planning with another donor (UNDP/UNCDF) to address such institutional development, and this is now serving as the basis for the Bank's follow-on project, the proposed Decentralized Planning and Financing Program (DPFP). The opening of water points in rural areas has undoubtedly relieved problems of clean water supply in many communities. Unfortunately, this has been somewhat offset by the fact that some water points have been polluted by floods and some pumps have broken down - overall, existing rural water points are tending to break down rather faster than new ones are created. This is the one area where community mobilization and participation were significantly strengthened by the project and this is helping to improve operation and maintenance. Perhaps the main impact of the water component has resulted from the recommendations of the Rural Water Supply study of 1997. These have led to the adoption of a demand-led water supply policy which has been institutionalized and is being pursued in the IDA-funded National Water Program. Based on the following assumptions: (i) schools: 2.5 shifts per school room; plus adult literacy of 400 places per school (3 km walking radius); and (ii) health: 5,000 patients per center with walking radius of 8 km. 5 The impact of the land mapping component has been minimal so far, perhaps because of relatively limited publicity about the existence and potential use of the new GIS and related maps which were created under the project. Consequently, these have only been used by DINAGECA and a few individuals who have purchased them in Maputo. They may in future contribute to land policy and law implementation, which is the responsibility of the Land Commission. 4.2 Output by Component: Component 1: Support to Decentralization The outputs from this component are substantially different from those envisaged in the SAR. The SAR placed emphasis on capacity-building through technical assistance at provincial level. It also foresaw using sub-projects to test procedures for planning, financing and implementing community-based initiatives and to improve knowledge and skills of local government officers, NGO staff and community leaders in decentralized community-based rural development. Sub-projects (up to US$3.6 million including community contributions) were to be specified by the communities, and were anticipated to be mostly related to production activities (such as agriculture). As designed, this component would have contributed to the NRP more in terms of institutional development than in the actual rehabilitation of physical infrastructure. As it turned out, relatively few production-related initiatives were pursued; much more infrastructure has been developed than was originally envisaged; but more limited progress than hoped for was made in engaging communities and local officers and establishing effective institutions for local governance through which they might contribute to the economic and social development of their neighborhoods. The project employed eight Rural Development Agents (ADRs), who worked as liaison agents between communities, districts and the provincial project units, but many micro-project proposals were initiated directly by district and provincial administrations rather than by communities themselves. Social Infrastructure: A total of 30 new health centers were built with 54 new houses for nurses; 93 schools were built with 24 new houses for teachers. The health centers and houses were built almost exclusively in Sofala, while 64 of the schools were built in Zambezia. In addition, 17 hospitals, health posts or centers and 25 schools were improved (rehabilitated, equipped or extended). Latrines were built at almost all new health posts and schools and at a number of similar additional public sites. Water points were also normally constructed with schools and health posts, as well as in residential areas (see also Component 4 below). Agricultural Development: As noted above, in contrast to expectations, less than 2% of micro-project funds were used for agricultural activities, mostly in Sofala. Livestock micro-projects, mainly goat re-stocking, were implemented in 6 districts of Sofala and one district of Zambezia, with the distribution, on a grant basis, of some 680 goats (over 200 beneficiaries) and 50 draught oxen (25 beneficiaries). Fruit tree, coconut and cashew planting material was provided to a few communities in Sofala. Three dip tanks were rehabilitated or constructed and two irrigation systems were assisted. The identification of these micro-projects was generally facilitated by ADRs, while formulation and implementation was done by the Provincial Directorate of Agriculture. Rural Market Infrastructure Development: Nine markets were rehabilitated and upgraded and 7 new markets were built. Eight maize mills were distributed and one rice mill developed. Over 200 km of feeder and access roads (including bridges and culverts) were 6 rehabilitated or reconstructed in Zambezia and 5 metallic bridges were erected in Sofala. These were particularly important developments in the relatively isolated rural economy of post-war Mozambique. The component's training program was late and spread much thinner than originally intended. Provincial and district officers, together with community and NGO representatives, were given a week's course in Micro-project Planning with Community Participation. Twenty-six courses were held, allowing all districts the opportunity for relevant officers to be trained. The courses, which included field training in 40 communities, were both well conducted and appreciated by participants. However, only 11 districts received a further course for consolidation, there was no sustained follow- up, and little subsequent use was made of the skills acquired by the participants in the implementation of the project. This was partly because of the frequent ltransfer of district staff. In terms of capacity- building, it is thought that project staff, including the ADRs, benefited most, largely through doing. Several are now using their skills and experience in sirmilar projects. The INDER training team became a well-established resource used by a number of other projects. Much less technical assistance (TA) was provided than that envisaged in the SAR. No internationally-recruited officers in planning and project appraisal were engaged for the provinces. With the notable exception of the second Project Coordinator in Zambezia, the managerial performance of the Provincial Project Managers was weak. The ADRs appear to have performed variably, but satisfactorily overall during the early stages of the project, considering the greatly extended areas they were expected to cover and the delay in the training program. However, most appear to have had little lasting impact on the practices of the district and local administrations. With the restructuring of the project (1997), the skills of the ADRs in extension and community development no longer fit with project implementation requirements. Proposals were made for their redeployment, but the ADRs seem to have been under-utilized in the latter stages of the project. The major achievements in system development and capacity-building have been in the sphere of public works construction, contracting and management.4 Following an initial period when construction was directly managed by the project, and both rates of progress and quality of works were found to be deficient, the provincial project units instituted a system of public bidding for project works which was open and transparent. As experience was gained through project implementation, an accredited list of contractors was prepared and, to reduce evaluation and administration work to manageable proportions, invited shortlists for individual works were often preferred. The project also instituted more effective and appropriate works supervision by appointing and instructing its own supervisors to ensure that workcs were continuously and promptly supervised, even in remote locations. Works were generally completed relatively rapidly and construction quality improved, although some defects such as warped roof'timbers, leaking roofs, damp walls and termite damage are observed in some places. The public work.s contracts provided job experience and on-the- job training to countless local workers, as well as 3 courses for nascent contractors and contractors' staff. A proposal was developed for the establishment of an independent agency (AGETIP) which could manage provincial public building programs across all sectors. The study was shelved but may be a useful input for the preparation of the follow-on project - DPFP. Overall, the outcome of this component is rated satisfactory. 4 See also a (draft) report by Louis Helling: The Policy and Institutional Framework for Decentralized Planning and Management of Small-Scale Public Infrastructure Investment for Rural Development in Mozambique. February 2000. 7 Component 2: Land This component was intended to contribute to national land management and land rights allocation systems. After project restructuring, the Land Use Mapping Component5 was comprised of several activities including: (i) the revision of national topographic maps and the estimation of land use patterns; (ii) the detailed mapping of priority RRP districts; (iii) the development of a national Geographic Information System (GIS); and (iv) training of DINAGECA staff. At project inception, INDER signed an implementation agreement with DINAGECA, which in turn contracted all the component activities to a joint venture between a specialized national and international company. A three-year contract was signed around mid-1996 and successfully implemented in two phases, generally in accordance with the agreed terms of reference and time frame. Each phase was concluded by a national workshop held in Maputo. The joint-venture contract included the procurement of equipment, software and satellite imagery. On-the-job training was provided to a team from DINAGECA and INDER. The feasibility of using 1:50,000 maps for community land demarcation, was also tested in one district in each of Sofala and Zambezia provinces. As planned, a GIS has been established within DINAGECA and the following maps (500 copies each) have been produced: (i) a national coverage of the land use map (1:250,000 scale) in digital format and as printed copies (93 sheets); and (ii) detailed land use coverage for 8 priority districts at 1:50,000 scale (88 sheets). The maps can be purchased at DINAGECA's office in Maputo. However, staff from the Provincial Service for Geography and Cadaster (SPGC) were not involved in the joint-venture contract implementation, they were not invited to the national workshops, and they did not benefit from any training: even field truthing was carried out by DINAGECA staff from Maputo. The maps and GIS, have remained rather confidential and seem largely under-utilized, hence the impact on addressing land management and land right issues has been minimal so far. Staff from SPGC are hardly aware of their availability, nor are they generally known to other provincial and district institutions and projects. For these reasons, the component is considered as marginally satisfactory. Component 3: Distribution of Agricultural Inputs (Agpacks) This component was designed as an emergency measure to support families returning to their farms after the end of the war. At the outset of the project, in early 1994, administrative problems6 meant that the opportunity was missed to procure Agpacks for the coming agricultural season. INDER submitted a procurement request to MPF in November 1994, but Government was not prepared to use IDA credit for this purpose and in 1996 requested the cancellation of the component. Funds were reallocated to other activities, particularly to Decentralization (micro-project sub- component). Component 4: Rural Water Supply At project restructuring, activities under the Land Study and Pilot Demarcation sub-component had not yet been initiated, and were transferred to ASRDP. 6 The Special Account was not opened until June 1994. 8 This component was planned to strengthen the capacity of 3 district workshops run by the Provincial Rural Water Supply Workshops (EPARs) and finance the construction of about I 00 water sources (by the workshops) per year. It was also designed to improve the organization and management of the EPARs and to develop a scheme for operating and maintaining water supply systems based on community participation and the provision of a standardized hand pump: the AFRIDEV. An inventory of water supply points in Sofala and Zambezia provinces was to be undertaken in the first year of the project as a basis for a 5-year development plan. During the early stages of project implementation, the role of EPARs was already being reviewed and the institutional strengthening, TA and training that were envisaged for them were not pursued apart from the supply of 6 tractors, 2 lorries and 2 vehicles. EPARs gradually began to operate on a commercial basis and competed for contracts with private companies. As a part of the restructuring of the Bank's portfolio, the ASRDP water component in Nampula was to be transferred to RRP. However, funds were not transferred from ASRDP to RRP and no reference to such a change in the RRP project area was made in the revised DCA. Hence, RRP took no action in Nampula. Water Point development. A total of 300 manual AFRIDEV pumps were supplied under the project, 150 in Zambezia and 150 in Sofala. Most pumps were installed by EPARs for micro-projects under Component 1, with civil works for the water points generally being undertaken by private contractors. Other water points were contracted to NGOs, other government agencies and private contractors. Almost all the pumps in Sofala have been installed in villages, while in Zambezia 100 pumps have been installed in villages, 45 in schools and 3 in health centers. The communities have the right of use and the responsibility to maintain the pumps whether the water point is in the village, school or health center. Community participation was promoted through the Community Education Program (PEC), a branch of EPAR, which was operating before RRP began. Consultation with the communities took place from the identification stage of the water point development process, but the approach was supply-based and did not critically review community demand. A new strategy is now in place which ensures a demand-led approach. Prior to construction of the water point, elected members of the community are trained in O&M, and I'EC officers continue to monitor the water points during the year after completion. Water points are, by and large, maintained by the community representatives, and some communities successfully raise funds for repairs and spare parts. However, some communities have been less inclined to contribute to O&M when the facility is located in public areas such as schools and health centers. In 1995, the project carried out an inventory of the location, condition and operation of each rural water source in Sofala and Zambezia provinces. A purpose-built database software was installed in the Water Departments of the Directorates of Public Works and in the EPARs of both provinces and training was given on the use of the software. The system has been operational in Zambezia since 1995 and data have been updated regularly. The resulting database has facilitated planning, O&M and emergency operations. The experience in Sofala has been disappointing. Most data and the software were lost when the computer broke-down. No attempt has been made to re- install the system and maintain the database - partly as a result of the frequent transfer of staff and lack of sufficient training. The project financed a national study on Institutional Arrangements for the Provision of Rural Water Supply and Sanitation (March 1997) which has laid the foundations for a national 9 rural water supply policy and strategy. The study's recommendations form the basis for the National Rural Water Transition Plan, providing guidelines and standards for the sector, including the future role of the private sector, and establishing a framework for improving the sustainability of water supply in Mozambique. The study also recommended a more thorough review of the mandate, legal status and role of EPARs. Small Piped Systems. Project restructuring allowed for the rehabilitation of small piped village water supply systems. Four feasibility studies were carried out for the towns of Gurue, lie, Maganja da Costa and Pebane in Zambezia province. The project completion date was extended to allow the completion of the Gurue system, which has now been rehabilitated and expanded. The rehabilitated system has been operating since September 2000, but some complementary civil works are required to protect the intake. Arrangements for the O&M of the scheme still have to be determined pending completion of a study by the National Water Department (DNA), which intends to use Guru6 as a pilot for a new organizational model. The overall outcome of this component is rated satisfactory. 4.3 Economic Rate of Return: An ERR was not calculated for the project. 4.4 Financial Rate of Return: (not calculated) 4.5 Institutional Development Impact: With the extinction of INDER, and its incorporation within MADER, the impact on institution-building for rural development remains to be seen. However, the project did not build significantly the capacity to undertake decentralized rural development, especially at district level. This was further exacerbated by the high turnover of district staff, including those who benefited from project training. The locus of decentralization policy and strategy development has moved to the Ministries of Planning and Finance and of State Administration. There will still, however, be a parallel need for local participation in planning for productive initiatives. Under the Rural Water Supply component, PEC teams have been strengthened, but with the commercialization of EPAR's activities there are uncertainties regarding PECs' operations. Under the Land component, capacity has been built within DINAGECA for operating GIS, but SPGCs did not benefit from such training as anticipated at appraisal. Overall, the Institutional Development impact is considered as being unsatisfactory. 5. MAJOR FACTORS AFFECTING IMPLEMENTATION 5.1 Factors Outside the Control of Government or the Implementing Agency: * Following the signature of the peace agreement, security in the country-side improved rapidly and provided favorable conditions for RRP operations, especially in the rural areas. * The loss of the first President of INDER, who was the prime mover for the borrower's conception, preparation, negotiation and early implementation of the project, negatively and substantially affected project implementation during 1995 and 1996. 10 * In 2000, a cyclone damaged rural infrastructure uncder construction in Sofala, which resulted in implementation delays and increased costs; during the same year, floods delayed completion of the Gurue piped system, and caused losses of oxen provided by one micro-project in Sofala. 5.2 Factors Generally Subject to Government Control: * The existence of INDER as a conduit for cross-sectoral investment funds made a substantial rural development program possible, during a period when the focus on sectoral programs created an environment in which sector-specific activities were receiving most attention and few other resources were available for cross-sectoral initiatives. * Government delayed appointing a successor to the first President of INDER until December 1996, and this caused a dramatic reduction in support and leadership at the highest level during 1995- 1996; on the other hand the appointment of the new President provided a new dynamism and contributed to improving project performance thereafter. * The unwillingness of Government to use IDA funds for international TA negatively affected project start-up. * There was poor understanding of the project's objectives and method of operation by some district administrations, which led to concentration of project implementation in certain districts and spheres of action. 5.3 Factors Generally Subject to Implementing Agency Control: * The initial low capacity of national and provincial jproject management, and their unfamiliarity with IDA procedures contributed to a rather slow project start-up. * However, continuity of the two key central project management staff, following a "learning-by- doing" process, resulted in a good performance by project management and this contributed to the successful implementation of the micro-project sub-component. * The failure to assure implementation of training arrangements on time resulted in some delay in community consultations and identification of micro-projects and may have prejudiced the ultimate outcome in terms of capacity-building. * In common with other Mozambican projects, there was an almost complete lack of M&E, which was critical for RRP, given the pilot nature of the project. Upon project completion, there is little information on systems for decentralization and community-driven development, cost effectiveness and project impact; as a result the project has been left in a condition in which it cannot completely fulfil its role as a pilot. 5.4 Costs and Financing: Total project costs (see Annex 2a) are the same as the SAR estimates (US$23 million including contingencies). Following project restructuring, the costs by component changed substantially, in particular for the Decentralization component which amounted to about US$16 million, as compared to appraisal estimates of US$8 million (and micro-projects amounted to some US$11 million compared to appraisal estimates of US$4 million). Bank disbursements amounted to some US$20 million, most of which were disbursed after 1996. The initial delays in disbursements were mainly due to lack of progress in implementation, in particular TA and cancellation of the Agpack component. Government contribution amounted to US$1.9 million, which is similar to the SAR estimates. Of the government funds, US$0.5 million were disbursed by the provincial governments of Sofala and Zambezia to micro-projects. Beneficiary contributions to micro-projects (planned in the SAR to amount to just over US$1 million) were never valued in 11 monetary terms. However, many communities did contribute to project costs by providing local construction material and labor. 6. PROJECT SUSTAINABILITY 6.1 Rationale for Sustainability Rating: Overall, project sustainability is rated as likely. Over 50% of total project costs were invested directly in rural rehabilitation and development through micro-projects, primarily for the construction and rehabilitation of schools, health centers, water points and roads. Their sustainability will largely depend on the quality of civil works and the capacity of the beneficiary communities, line ministries, and provincial and district authorities to ensure their maintenance. The high rate of use of these infrastructures, as observed by the mission, suggests a strong interest by stakeholders in their continuing operation and maintenance. The proposed follow-on project, the DPFP, is geared to strengthening arrangements for O&M of rural infrastructure. Although there will be a time-lag between the two projects, the sustainability prospects of RRP investments through the micro-projects would be enhanced if DPFP continued to cover the provinces of Zambezia and Sofala. Regarding the Land Use Component, DINAGECA has a team of trained GIS and remote sensing specialists who have the capacity to maintain the system. However, equipment will rapidly become obsolete, software will need upgrading, and the GIS will have to be updated. DINAGECA, which does not have the financial capacity to sustain and further develop its GIS, needs to seek further external support to ensure the sustainability of its GIS-related operations. If it is ultimately decided to privatise the EPARs, special arrangements will be needed to ensure that the community work now carried out by the PEC is continued. 6.2 Transition Arrangements to Regular Operations: The project management teams in Maputo and the two provinces are being retrenched, and project assets and files handed over to the relevant institutions at national, provincial and district levels. Responsibilities for future operations regarding micro-projects, land use and rural water supply will rest with the respective line ministries. DINAGECA is committed to distributing the printed land use maps and to organizing workshops to train stakeholders on their use as well as the potential uses of digitized maps/GIS information. Such training is to involve staff from SPGC, districts, NGOs and development projects. In the longer term, the GIS developed under RRP should be integrated within a long term strategy for land information management, for selective land titling and cadaster purposes and for monitoring land use and management (e.g. changes in land use or land cover, agricultural statistics, forest, environmental aspects etc.). This integration could be undertaken by the National Land Commission and involve the key players such as DINAGECA, SPGCs, PROAGRI, the environment and forest departments (MICOA, DNFFB) and other related programs. DNDR should liaise with DNA and the World Bank National Water Project to ensure that final works of Gurue are completed and O&M arrangements are put in place. 12 In 1999, the project commissioned a study entitled "The RRP Experience with Decentralized Management: Lessons for the Future". The draft report provides useful information, but the study remains incomplete, especially regarding the civil works aspects. This study should be revisited and completed in the context of the proposed DPFP formulation, and a thorough review of civil works construction7 should be made. The review should be expanded to cover similar facilities constructed by other agencies. As part of the same exercise, the proposals made in the studies relating to the establishment of provincial AGETIP entities should be critically reviewed. It would also be helpful to review in detail the experience of the project's ADRs in working with local communities and providing liaison with district and provincial level entities. 7. BANK AND BORROWER PERFORMANCE Bank 7.1 Lending: The Bank's lending performance is rated unsatisfactory, primarily because: - the SAR was not fully in harmony with Government's priorities for national reconstruction, and Government commitment to the project design and financing arrangements was not properly assessed; * the comprehensive preparation work (e.g. on training) was not adequately reflected or built on in the SAR; * the Bank did not ensure consistency between SAR, DCA and the project's Operations Manual (in particular with regard to social infrastructure); * project design was complex; and * the large amount of the credit was inconsistent with a pilot project and the SAR disbursement schedule was manifestly unrealistic. 7.2 Supervision: The Bank's supervision performance is rated satisfactory overall. Despite the early diagnosis that INDER would need strong guidance ancd support, the Bank did not provide the benefit of its experience elsewhere to the project. There was a gap in supervision between April 1994 (when the project was rated satisfactory) and May 1995 (when the project was rated unsatisfactory). No joint annual implementation reviews were carried out as agreed upon at Appraisal. After 1996, there was continuity of TTL, detailed field supervision missions and implementation reviews were conducted, guidance was timely and effective, and good working relations with the borrower were maintained. The supervision performance during this period contrilbuted substantially to the success of the project and is rated highly satisfactory. The mid-term review was very helpful in turning around what had been a shaky initial period - this despite shortcomings of the original design. 7.3 Overall Bank Performance: To review the construction design, construction costs, quality of works and related maintenance costs, tendering and supervision arrangements, etc. 13 In spite of the shortcomings during lending and supervision until 1996, the overall Bank performance is considered satisfactory, primarily because of the key contribution of supervision to satisfactory project implementation during the last four years of the project. Borrower 7.4 Preparation: The Government was actively involved in the project preparation, and the first President of INDER contributed directly to the process. However, Government should have been more pro- active during appraisal to ensure a better balance between piloting decentralization procedures and contributing to the National Reconstruction Program in terms of rural rehabilitation. 7.5 Government Implementation Performance. Throughout implementation, Government commitment to the project was strong. Counterpart funding was made available although it lagged at some stages of the project. Audits were carried out regularly but failed to discover a systematic error in withdrawal applications. 7.6 Implementing Agency. INDER gave high priority to the implementation of this project. In part this was due to the strong interest in the program held by the first and later the second President of INDER. The importance of the President's interest in the program was particularly evident following the loss of the first President of INDER, when the project team did not receive adequate support. The direct personal interest and practical support of the second President of INDER and continuity of project management have been key in this regard. Innovative solutions have been found to resolve the problems that have arisen during implementation. 7. 7 Overall Borrower Performance Project administration has been adequate overall, although some lapses, both in the provinces and at headquarters did occur. The project management team has been less satisfactory in reporting, and in developing an effective monitoring and evaluation system, which have been minimal. Overall, the borrower's performance has been satisfactory. 8. LESSONS LEARNED In the course of project implementation, the following lessons were learned: * The SAR, together with the Operations Manual, rapidly proved to be an inadequate guide for project implementation. At Appraisal, the Bank should ensure clarity of purpose and agree on these with the relevant parts of Government. This is especially important where the weight of preparation has fallen predominantly on Bank staff and consultants, rather than Government itself. 14 * The IDA Credit included funding for massive TA and emergency assistance (Agpack component which included provision for contracting NGOs). During project implementation, it appeared that Government was not prepared to use IDA funds for such type of expenditure. Government should seek a blend of credit and grant funding through co-financing arrangements for operations that include emergency assistance or require substantial TA and a partnership with NGOs. The Bank should also help borrowers in seeking grant co-financing. * The project activities and outputs are substantially different from those anticipated at appraisal, which recognised the need for flexibility in a post war situation. Managed flexibility is the key to the success of project implementation, in particular when the situation is expected to change at a fast pace. * The RRP was designed as a three-year pilot operation and it was anticipated that much of the US$20 million Credit would be spent over that period. The project ended up as a 7-year rural development operation. Pilot activities should be limited in size and an effective M&E system, backed by targeted Bank supervision, should be in place to "capture" the lessons learned for future larger scale operations. Should M&E capacity of the implementing agency be weak, it could be contracted to qualified national institutions or the private sector. * Following restructuring, the project was significantly simplified and much of project management attention was devoted to the micro-project sub-component, which proved to be successful. Micro-projects were focused primarily on the rehabilitation and development of social and rural infrastructure. Simple and targeted activities greatly contribute to the success of such operations. - INDER and some key project staffiwere committed to project success and achieved much, despite relative inexperience and being few in number. Where the will to succeed and freedom of action are present, national staff can successfully implement projects with minimal outside assistance. * Micro-project success and sustainability is more likely when communities are involved throughout the process, from the identification and design stages to completion, operation and maintenance. * Addressing generic problems observed on civil works (e.g. roof water leakage, moisture infiltration and termite infestation) would not require major incremental costs at construction, but they would reduce the maintenance cost and improve sustainability. * Land use mapping is not an end in itself, but rather a tool for supporting the implementation of the land policy and law, as well as for land/natural resource management purposes. Project design should ensure that such an expensive tool be part of a comprehensive strategy for land policy development and management. 15 9. PARTNER COMMENTS (a) Borrower/implementing agency. (attached as Annex 8) (b) Cofinanciers: NA (c) Other partners (NGOs/private sector): NA 16 10. ADDITIONAL INFORMATION Annex 1 Key Performance Indicators Output Indicators: I l l Indicator pro 2 In AptuaULatest Remarks _________________________ .tSAW AP I EsUmite Support for Decentralization Micro-Projects implemented: Zambezia No. of new schools constructed Not defined 71 No. of schools rehabilitated Not defined 22 Furnishing and other equipment for Not defined 87 schools No. of houses built for nurses and Not defined 11 -teachers _No. of health posts constructed Not defined 2 _ Rehabilitation of health centers Not defined 12 No. of latrines constructed Not defined 364 Micro-Projects implemented: Sofala __ __ No. of new schools constructed Not defined 91 No. of schools rehabilitated Not defined 8 Furnishing and other equipment for Not defined 32 schools No. of houses built for nurses and Not defined 76 _teachers No. of health posts constructed Not defined 30 Furnishing and other equipment for health Not defined 5 _centers IRehabilitation of health centers - Sofala Not defined 5 _No. of latrines constructed Not defined 236 17 Output Indicators: 2 ;1 ' ' ;-*i -dicaito i - 0 . Remiarlks Su Ft fr JDeeetrliato ____X____X____ Land Study: Land use, ownership, social 4 Studies Was transferred to the issues and conflict solving. (2-Sofala and Not Done ASRDP-MADER. 2-Zambezia) Produce a report on demarcation By Project Year Partially Done A clear methodology experience with methodology 3 and guidelines have not recommendations on smallholder been produced as demarcation intended National topographic and land use map X 93 sheets Done 500 colour copies, scale (1:250,000) Produce detailed land use maps X 500 for 88 sheets Done 4 priority districts, 2 in Sofala and 2 in Zambezia (at 1:50,000 scale) GIS: supply software, manuals, compile Set Complete data and train DINAGECA staff in the use of GIS. Long-term training (university degree) 2 DINAGECA Not Done _______ _______ _______ _______ ______staff 200,000 - Cancelled at Distribution of a 'iousholds _____________ restructuring R r Water uup 0 lWyt =X'_ _ Establish District Workshop in Buzi, Sofala 1 Cancelled at _ _restructurin Establish Workshop in Chinde & Alto 1 Cancelled at _Molocue, Zambezia restructuring Provide equipment to workshops to-increase capacity, - Cancelled at logistics and transportation. restructuring IConstruct Water Sources in Sofala Not defined 95 Construct Water Sources in Zambezia Not defined 52 No. of Pumps supplied 1000 300 Inventory of Rural water supply in Project Year Complete for Operational in Zambezia infrastructure for Sofala and Zambezia 1 Sofala and but not in Sofala _________________________________ _________I Zambezia 1/ The SAR had not defined any quantifiable impact or outputs for the RRP. 18 Annex 2a Project Costs and Financing Project Costs by Component (in US$ million Equivalent) Proiect ComDonent A_ raisal Actual/Latest Percentage of Estimate 1 Eftfmate /2 Appraisal Support to Decentralization 7.9 15.8 199.2 Courses and Seminars 0.7 0.4 51.4 Sub-Projects 4.3 10.8 250.5 Capacity Building /3 2.9 4.7 160.3 Land 3.5 3.4 96.4 I Land use Mapping & Inst. 2.7 3.3 121.5 Study Land Study & Pilot 0.9 0.2 17.5 Demarcation Distribution of AgPacks 5.4 - - Rural Water Supply 6.2 2.4 39.5 t Sofala 1.9 0.5 23.4 Zambezia 3.2 0.6 84.3 Small-Piped System _ 0.9 - Support to Center 1.0 0.5 48.0 Total: 23.0 21.7 94.1 1/ Mission estimates (the SAR did not present costs by component with contingencies). 2/ Mission estimates (project accounts did not keep records of expenditure by component). 3/ Includes project management, consultancies and studies, see supplementary Table 3. 1'9 Annex 2b Project Costs by Procurement Arrangements (in US$ Million Equivalent) Expenditure Categories Procurement Method Appraisal Procurement Method Actual/Latest Estimate - 0 Estimate ICSB NC8 Other NBF Total ICB NCB Other NBF Total 1-A Civil Works for Rural 2.2 2.2 0.7 - 0.0 - 0.7 Water Supply 1-B Building' t Reoation - - 0.3 0.3 - - 0.4 - 0.4 (houses & offices) 2 Sub-Projects(A.2) 0.3 0.7 1.5 1.1 3.6 - 9.1 2.0 11.0 3-A Vehicles & Equipment 2.3 - 0.4 2.7 1.1 - 0.6 1.7 3-B Seeds & Tools 4.7 - 0.1 - 4.8 - - - 4-A Land Use Mapping 1.4 - - 1.4 3.2 - 0.1 3.3 Contracts 4-B Technical Assistancei - : 4.8 - 4.8 0.6 1.6 0.7 2.9 4-C Training 0.9 - 0.9 - - 0.4 0.4 4-D NGO Services 0.8 - 0.8 - 0.0 - 0.02 5 Incremental Operating - 1.5 - 1.5 0.0 1.4 - 1.4 Costs Total 8.7 0.7 12.5 1.1 23.0 5.6 10.7 5.4 - 21.7 11 Technical assistance includes Project Preparatory Fund (PPF) refinancing. 20 Annex 2c Project Financing by Category of Expenditure (in USS million equivalent) Cat Cateqory of Expenditure Appraisal Estimate Actual/Latest Estimate Percentage of Appraisal No. Bank Govern Beneficia ank 12 Govern Beneficia Bank Government Benefi ment ries ment ries 13 ciaries |1-A Civil Works for Rural 1.9 0.2 0.1 0.7 0.1 - 34.7 29.0 - Water Supply 1-B Building Renovation 0.2 0.1 - 0.1 0.2 - 55.9 248.5 - (houses & offices) 2 Sub-Projects (A.2) 2.5 0 1 1.0 10.2 0.8 - 407.8 847.7 - 3-A Vehicles & Equipment 2. - 1.2 0.5 - 45.0 - - 3-BSeeds & Tools 4.8 4- Land Use Mapping 1.4 - - 3.2 0.1 - 227.9 - - Contracts 4 B Technical Assistance 1i 3.4 1.4 - 2.8 0.0 - 83.6 1.1 - 4-C Training 0.9 - 0.3 0.0 - 38.3 - - 4-D NGO Services 0.8 - 0.0 - - 2.3 - 5lncrementalOperating 1.4 0.1 - 1.1 0.2 - 81.2 222.6 Costs | Total 20.0 1.9 1.1 19.7 1.9 - 98.6 100 11/ Includes PPF. l l l l _ 2/ The Bank's latest disbursements figures are US$300,000 less than the Project accounts. 3/ Beneficiary contribution was mainly in terms of construction material and labor, no attempt has been made to value these inputs. Annex 3 Cost Benefit Analysis (Indicate Currency, units and base year) Present Value of Flows Economic Analysis Financial Analysis _____ Appraisal [Latest Estimates Appraisal Latest Estimates Benefits An ERR was not calculated for the project Costs Net Benefits IRRINPV 21 Annex 4a Bank Inputs: Site Visits Stage of Project Cycle MonthNear ~No of perso.: Perf a i i :0: ia S0:0 i t ; I p:ementat:Ion d t3~elo .t: Identification 1/ May-June 90 EC (2), Ag, nla nJa R.Eng, Ext Preparation 1/ Apr-92 EC (3), RD, Ag, n/a n/a WID Appraisal/Negotiations Jun-July 92 EC, RC, IS, n/a nia L.Info, FS, H- Eng Svp.rvi.sions :00dSPNl.10 7 7 0 7 0 May-June 93 L-lnfo, LM, EC, S S ME SPN-2 Sept-Oct 93 EC (3), Ext, S S AGME, LM SPN-3 January 94 EC, Ag S S SPN-4 March 94 Ag, RW, EC S S SfP-s December94 EC, CP U U PN-- July 95 EC, L.Info, CP, U U ME 70 SP '. : 00 00:T0December 95 EC U U SPNW8 April-May 96 EC U U SPN- O-Oct-Nov 96 EC, C.Eng, S U L.Info SPW-IO June-July 97 EC, L.lnfo, S U C.Eng, W.Eng SPN-i1 Oct-97 W.Eng S S SPW!2 November 97 EC S S SPN-13 April-May 98 EC (2), Env S S _P -14 Oct-Nov 98 EC (2), C.Eng S S 7 - Jn June 99 EC S S S0pNs18 Oct-Nov 99 EC S S 7SPN1 7 Feb-Mar 00 EC S S 0 1N-48 May-June 00 EC S S SPN-T1 Nov-Dec 00 EC S 7770 XICR Jan-Feb 01 Ag, EC (2) S S 1/. Including contribution from FAO/CP: Notes: Key to Specialized staff skills Institutional Specialist IS Land Information Specialist L lnfo EconomistUAgrc. Economist EC Food Security Specialist FS 22 Stage of Project Cycle I Month/Year No. of Persons Performance Radings and Speciality Implementatton Development IProgress Objectives Agronomist/Agriculturalist Ag Hydraulic H-Eng R I Enineer R.Eng Engineer L Rural Engineer R.Eng Land LM l___________________ Management Agriculture Extension Sp. Ext. Rural Water Specialist RW WID Specialist WID Community Participation Specialist CP Rural Development Specialist RD Monitoring & Evaluation Specialist ME Rural Credit RC Civil Engineer C-Eng. Water Engineer W.Eng Ratings: U= Unsatisfactory S= Satisfactory Annex 4b Bank Inputs: Staff * Stage of Project Cycle Actualf Laest Estimates ___ ____ Staff Weeks US$ (000G Identification through 118.4 312.2 Negotiation 1/ Supervision 131.8 406.5 ICR to be completed by AFTR1 to be completed by AFTR1 Total . | 250.2 718.7 1/ FAO/CP and Trust Fund contributions at identification/preparation stages are not included. 23 Annex 5 Ratings for Achievement of Objectives ** / Outputs by Components Objectivelutputs Highly tItuac6toy atisfawtory ifghly Not ___________: f ;:; C . Unsatisfacory Applkable MoiroPoies t007 07:88:::: _.X x~~~~~~~~~ x~~~~~~~~~ x~~~ fW0000** Re--lates toffi the; ob0jectives spcfe Xi hA 01~~~Rlte toS the00 outputs spcfe in... th P 0ADX\ Annex 6 E7iW~~~~~Rtig ofL Ban and Boroe PerformanceL PubEicSe;Isatisfactor ______ Ustsatr Banherf(Spoirmc: .e,in ______ _____ ________ Notes: * Also includes Bank-financed and trust fund consultants j ** ~~Relates to the objectives specified in the SAR r-Spriin _____ ______I_____ __ r *** ~Relates to the outputs specified in the PAD Annex 6 Ratings of Bank and Borrower Performance lmplementation ~ ~ ~ ~ ~ ~ Jusiofctr Bank Performance: 7~~~~~~~~~~~~ Borrower Performance: Go:;4vernment X lmIlementing tAtency X Overall ___________ X:::OveiX _ 24 Supplementary Table 1 Proiect Costs by Expenditure Category (in US$ million equivalent Expenditure Cateaories AchlaIsaI Acua Percentaate of Estfrnate Esftate A ralsal 1-A Civl Works for Rural Water 2.2 0.7 33.2 Supply 1-B Building Renovation (houses 0.3 0.4 120.1 & offices) 2 Sub-Projects (A.2) 3.6 11.0 306.9 3-A Vehicles & Equipment 2.7 1.7 61.8 3-B Seeds & Tools 4.8 4-A Land Use Mapping Contracts 1.4 3.3 233.4 21 4-B Technical Assistancell 4.8 2.9 59.5 4-C Training 0.9 0.4 39.4 4-0 NGO Services 0.8 0.0 2.3 5 Incremental Operating Costs 1.5 1.4 90.6 - Total 23.0 21.7 94.2 1/ Includes Project Preparation Fund (PPF). I _ 2/ The high increase in this category is partly due to lthe inclusion of Technical Assistance |within the joint-venture contract (rather than under category 4B, see also Annex 2a). '25 Supplementary Table 2 Proiect Costs by Cateaorv of Expenditure (in SDR million equivalent) 0 E d S i 0; Es0lmat ;f of AAiur a lasal I Works 1-A Rural Water 1.09 1.09 0.71 0.55 0.50 45.83 1-B Other 0.07 0.07 0.08 0.08 0.08 116.06 2 Sub-Projects (A.2) 1.66 6.03 7.56 7.56 7.38 444.74 3 Goods 3-A Vehicles & Equipment 1.44 0.98 0.88 0.84 0.83 57.84 3-B Seeds and Tools 3.25 - - - - 4 Consultant Services 4-A Mapping 0.72 2.20 2.28 2.28 2.31 321.16 4-B Technical Assistance 2.21 1.81 1.96 1.98 2.04 92.18 4-C Training 0.36 0.36 0.23 0.27 0.25 68.73 4-D NGO Services 0.58 0.03 0.01 0.01 0.01 2.20 5 Operating Costs 0.80 0.80 0.79 0.93 0.81 101.07 Unallocated 2.32 1.13 - Total: 14.50 14.50 14.50 14.50 14.21 98.03 Source: From the WB disbursement schedule (February 2001). 26 r_ _ 1Supplementary Table 3 Decentralization Component - Capacity Building Sub-Component: Capacity Building 1/ Value Civil Works for INDER and other 328,0001 Long term Consultancy Training 72,000 =Auditing 219,000 NGO Services 18,000 O Costs 1,342,000 Project Staff Center 2/ 578,000 _Provinces 676,000 Short Term Consultancies Studies Beneficiary Assessment 39,000 AGETIP study (Aggregate) 337,000 Manual of Procedures 33,000 Training Plan/Manual 16,000 Decentralized Planning Report 15,500 Other 17,500 Vehicles and Equipment 611,000 |Training 3/ 1 376,000 Sub-Total: 1 4,678,000 1/ Refers to Annex 2a, Decentralization component. 2/ Including project director costs over 8 years (US$:343,000) Supplementar.y Table 4 I Micro-Proiect Sub-component: Number of Contracts completed by Type and Value 1/ Micro-Project Completion Date 1995-96 1997-98 1999-00 Total Zambezia: L Force Account No. 13 41 4 58 Total Value (US$ '000) 130.0 432.4 70.7 633 Sub-Contracting No. 201 103 227 350 Total Value (US$'000) 197.7 1,063.4 3,161.2 4,422 Sub-total Zambezia (US$ '000) 32. 1,496 3,232 5,055 Sofala: Sub-total Sofala (US$'000) 1,139.2 ~ 2,674.71 2,161.51 5,975 Grand Total (US$ '000) 1,466.9 4,170.5j 5,393.31 11,031 1/ Refers to Annex 2a, Decentralization Component/Micro-Project Sub-Component. 27 Supplementary Table 5 Decentralisation Comronent - Courses and Seminar Sub-Cornonent: Detailed Costs (Latest Estimates) No. of Participants Course/Seminar Title SAR Actual/ Latest Estimates Basic Course on Project Proposals 40 Capacity Building Contractors 71 Training for Supervisors I 83 Community Training in Participatory Planning 257 Income-Generating Project Planning Course (ADRs) 14 Topography Course, DINAGECA staff, South Africa 2 O&M of Buildings 49 Training of Community Leaders in Rural Participation 60 - Rural Diagnosis for community land demarcation 11 WB-Seminar on procurement (Maputo/Lisbon) 5 Seminar on Rural Water Sector Strategy Study 40 Policy Seminar for Government officials 45 Accounting and Project Proposal Seminars 45 5 Micro-finance seminar, WB-Washington. 1 Study Tour on Micro Projects (Gaza Province) 5 Study Tour, Namibia 3 Studies M. Months Mapping 24 Outsourced by output XLand Tenure Study 24 Cancelled at I restructuring Inventory of Rural Water Supply 3!Outsourced by output 28 Supplementary Table 6 Technical Assistance: Comnparison between SAR and Actual SAR Actual/ Estimates Latest No. of Person Months Project-Financed National Staff Project Director 36 98 Financial Controller 36 72 Financial and Accounting Officers 36 212 Rural Development Agents (ADR) 36 488 Training Coordinator 36 0 Project Coordinators (Sofala, Zambezia) 115 Construction Engineers (Sofala, Zambezia) 96 _ __ _ _ _ __ _ _ __ _ _ _____ l I International Consultants: Program Advisor 20 0 Project Planning and Appraisal Specialist 36 0 Training Advisor 0 12 Water Supply Specialist (UNV) 36 0 Financial Management Specialist 18 0 Procurement Advisor 6 0 Legal Advisor 12 0 Public Administration Specialist 12 0 Monitoring and Evaluation Specialist 6 0 Specialist Trainers 12 0 Land Use Consultant 12 0 Community Education Specialist (Sofala & Zambezia) 4.5 0 Management Consultant 12 0 _ Management and Organization Specialist 6 _ National Consultants: Course Director 60 1! Course Lecturers 60 _ - Specialist Trainers _ _ 1/ Performed by INDER/DNDR regular staff. 29 Annex 8 GOVERNMENT OF MOZAMBIQUE IMPLEMENTATION COMPLETION REPORT MINISTRY OF AGRICULTURE AND RURAL DEVELOPMENT NATIONAL RURAL DEVELOPMENT OFFICE RURAL REHABILITATION PROJECT (Credit 2479-MOZ) Government Contribution to the Implementation Completion Report Project Data Identification February 1990 Project Concept Development May 1991 Pre-appraisal September 1991 Project Preparation Facility (PPF) Advance May 1992 Appraisal June 29-July 17, 1992 Approval by World Bank February 1993 Negotiations February 16-19, 1993 Signature of Credit Agreement April 30, 1993 Ratification by Council of Ministers July 16, 1993 Concurrence of Legal Counsel October 19, 1993 Credit disbursement December 8, 1993 Amount of Credit US$20,000,000 Government's contribution US$2,000,000 Community's contribution US$1,000,000 Projected closing December 31, 1998 Actual closing December 31, 2000 Coordinating institution Rural Development Institute (INDER), fol- lowed by Agriculture Ministry (MADER) Implementing agencies National Office of Geography and Mapping (DINAGECA)/Provincial Offices of Planning and Finance of Sofala and Zamb6zia provinces/National Water Office (DNA)IPRONAR INITIAL OBJECTIVES The specific objectives of the project were: (i) to strengthen the capacity of the provincial institutions responsible for planning, implementation, and management of national reconstruction, and to test and improve procedures for decentralized rural development; (ii) to 30 compile data on land use to guide future infrastructure investment planning, resettlement of returnees, and development of a land policy guaranl:eeing security of land tenure for farmers; (iii) to help farmers returning to their original landholdings to regain self-sufficiency in food production, thereby reducing the need for imported food and restoring the country's productive capacity; and (iv) to improve living and health conclitions in rural communities by providing clean water, reducing the time and energy put by women and children into fetching water, and increasing opportunities for productivity in labor, education, and other economic and social activities. INITIAL COMPONENTS OF THE PROJECT To achieve these objectives, the project was di\ided into the following four components: decentralization (which included training); rural water supply; land development (which included mapping and soil studies); and resettlement. Details of the contents of each component are given in the Staff Appraisal Report (SAR). The project, designed for implementation over Five years with closing projected for December 31, 1998, was intended to cover five districts in Zambezia province (Upper Mol6cue, Chinde, Gurue, Mocuba, and Nicoadala). and three districts in Sofala province (Buzi, Dondo, and Nhamatanda). These districts were chosen on the basis of security and highway or water access, given the circumstances of the armed conflict. This area of coverage was subsequently expanded to encompass Zambezia and Sofala provinces in their entirety, a total of 28 districts - a positive change because it allowed promotion of rural development in areas of greater need than those originally envisioned. Operation of the project in all districts of the two provinces became possible as a result of signature of the Peace Accord in 1992, without which rehabilitation and construction activities in these districts would have been impossible. This design modification showed that both the Government and World Bank were able to exercise flexibility in the interests of projec-t implementation, while the resulting expansion of coverage made the geographical scope of the operation more satisfactory. The project was coordinated by the Rural Development Institute (INDER) until 1999, and subsequently by the National Directorate for Rural Development (DNDR) of the Ministry of Agriculture and Rural Development until its completion on December 31, 2000. The implementing agencies were: the Provincial Offices of Planning and Finance of Sofala and Zambezia, for the decentralization component; the National Office of Geography and Cadastre, for the Cadastre component; and the National Water Office, for the rural water supply component. Although the project had been designed as a pilot scheme, its results proved to be those of a full-scale rural development project. It was carried out during a period of rapid changes in the country's economy and political situation associated with the postwar national reconstruction process. Constant readjustments were needed to ensure ongoing coherence with the objectives and policies of the National Reconstruction and Development Program. Amendments to the Credit Agreement After project implementation began, the Credit Agreement was first amended in 1996, its objectives being redefined as follows: (i) to strengthen the capacity of the institutions responsible for postwar reconstruction and to develop procedures for decentralized rural development; (ii) to compile data on land use and occupancy; and (iii) to improve the health :1I and living standards of rural communities by expanding rural water supply coverage. The main features of this amendment were: inclusion of funding for water supply systems, cancellation of the resettlement component, and cancellation of the sub-component on soil studies. The following objectives were set: Decentralization Component (a) Provide technical assistance for the National Rural Development Institute (INDER) and the key project implementing agencies, (b) identify, design, appraise, and execute small- and medium-scale investments in Sofala and Zambezia provinces as a pilot scheme, and (c) organize training courses and seminars. Mapping Component (a) Update the national topographic map and estimate actual land occupancy on a scale of 1:250,000, (b) prepare more detailed maps on a scale of 1:50,000 for eight selected districts, and (c) prepare the Geographic Information System Manual, and train technicians in operation of the GIS. Rural Water Supply Component (a) Rehabilitate and increase water supply capacity, mainly by (i) developing additional capacity, (ii) utilizing existing production capacity better, and (iii) rehabilitating or constructing wells, cisterns, and small water supply systems. (b) Expand the service delivery area by improving execution capacity, mainly through (i) developing procurement procedures for use by private companies in-carrying out projects, (ii) reducing the structural costs of rural water companies through improvements in their management practices that will enable them to compete with the private sector, and (iii) establishing a planning and monitoring system. (c) Develop an operation and maintenance methodology for water supply systems, based on community participation and such simple technical moves as (i) replacing existing hand pumps with others that are easier to operate and maintain, and (ii) drawing up an inventory of existing water supply infrastructure, with indications of location, condition, and need for repair. Other Amendments to the Credit Agreement The second amendment redistributed the available funds among eligible expenditure categories and extended Credit duration to December 31, 1999. The third amendment redistributed the available funds among eligible expenditure categories and extended Credit duration to June of 2000. The final amendment merely extended the duration further to December 31, 2000. None of the amendments changed the amount of the initial Credit Agreement. The amendments to the Credit Agreement-although made without a total revision of the initial document and the initial objectives of the project-were a sensible reaction to the changes that occurred in the process of.national, post-conflict reconstruction. They were further confirmation of the previously noted flexibility where implementation of the project was concerned, simplifying it and expediting disbursements. 32 OUTCOME/ACHIEVEMENT OF OBJECTIVES Decentralization Component In the rural zones, the decentralized pilot pro ject attempted to (i) test and improve procedures for the financing and implementation of small projects to benefit communities, support a resurgence of economic activities, improve delivery of social services. generate jobs. increase output productivity, construct and revive economic and social infrastructure, and finance small projects; and (ii) build up the capacity of the government agencies responsible for identifying, planning, carrying out, and evaluating national reconstruction and rural development activities projects, and programs, by using the procedures outlined above, doing so in a decentralized fashion and through training and instruction. Implementation of this component led to the signing of 667 contracts, which resulted in the execution that many micro-projects distributed among the various categories listed in the following table: 33 Programmed in the SAR Accomplished Type of Micro-project Unit Sofala Zamb. Total Sofala Zamb. Total Road repair Km 40 12 52 3 201 204 Bridge and aqueduct construction No - - - 6 25 31 Irrigation systems Hect 150 440 590 60 100 160 Repair of markets N 20 20 40 1 1 1 Reconstruction of markets N I1 11 22 3 7 10 Construction of Markets N 12 12 24 2 6 8 Construction of vendors' stands No - - - 335 440 775 Construction of elementary schools No - - - 24 68 92 Remodeling of elementary schools No - - - 3 22 25 Construction of secondary schools No - - - 2 - 2 Remodeling of secondary schools No - - - 2 - 2 Classrooms for 500 students each No - - - 104 250 354 Construction of 2-seat student desks No - - - 2.600 6.250 8.850 Expansion of Health Centers No - - - 3 1 4 Construction of Health Centers No - - - 26 1 27 Remodeling of Health Centers No - - - 5 5 10 Sets of hospital furniture and equip. No - - - 1 5 6 Construction of teachers' homes No - - - 26 - 26 Construction of nurses' homes No - - - 47 8 55 Construction of INDER offices No - - - - - 1 Bathrooms (schools and health cntrs.) No - - - 222 483 705 Construction of water points No - - - 55 32 87 Repairing of water points No - - - 40 20 60 Development of draft animals Cattle - - - 50 - 50 Development of goats Goats - - - 760 32 792 Fishing kits No - - - 6 - 6 Development of bee culture Hives - - - 18 - 18 Development of cashews Seedlings - - 24.000 - 24.000 Development of coconut Seedlings - - 6.100 - 6.100 Development of fruit trees Seedlings - - 4.087 - 4.087 Construction of dipping tanks Tanks - - - 4 - 4 Rice hulling machine No - - - I - I Corn mills No - - - 3 5 8 The 667 contracts made it possible to carry out micro-projects for reconstruction, expansion, new construction, and fitting out of public service buildings. These micro-projects were executed directly by the project in the first phase, and subsequently by local contractors. They were selected by the communities and the district and provincial authorities in the respective sectors. They were carried out in coordination with the provincial authorities whenever the micro-projects to be executed involved increased government spending for their operation. Taking into account that funding of the project was part of the National Reconstruction Program, the results for this component surpassed those anticipated in the initial project document. The micro-projects carried out contributed to rural development, particularly in 34 access to education, health, water supply, roads, farm output, and employment (albeit temporary). Health, education, agriculture, and water supply were the areas receiving greatest impact, as can be seen in the following table. Nearly 50,000 students, 2,600 families, and 960,860 people benefited from project activities in this component that financed micro-projects. Number of Beneficiaries Class 2 Sofala Zambezia Grand Total Students Classrooms 16,600 28,506 45,106 Subtotal - Students 16,600 28,506 45,106 Families Draft cattle 25 25 Breeding goat;s_9_ l 10 205 Coconut seedlings 1,725 1,725 Various fruit tree seedlings 488 488 Mandioc seedlings 50 50 Nurses' residences _48 48 Guard's residence I I Manager's residence I I Nurses' residences _ 47 8 55 Rural A.D.'s residence 1 I Teacher's residences 28 28 Subtotal - Families 2,51_2 18 2,530 Persons Rural markets 86,500 141,168 227,668 Health Centers 203,493 9,200 212,693 Rice hullers 42 42 Fishing kits 6 6 Traditional midwives trained 3,546 3,546 Wells for drinking water 51,000 28,500 79,500 Rural hospitals 25,000 39,752 64,752 Sets of hospital furniture 6,000 6,000 Kilometers of roads 78,247 78,247 Sewing machines 10 1 0 Mills 16,000 37,453 53,453 Bridges and aqueducts 172,534 172,534 Clinics 7,000 10,766 17,766 Cholera prevention 28,688 28,688 Hectares irrigated 30 14 44 Public sanitation facilities 7,440 8,548 15,988 Total - People 394,993 565,868 960,861 This component called for training courses and seminars to facilitate project implementation and prepare national reconstruction projects, as well as to provide training for personnel involved in regional planning in the participating agencies. Activities accomplished in this subcomponent can be summarized as follows: 35' Training and Instruction Type Expected Actual number of number participants Courses Basic Course on Projects for Project Staff 40 0 Training Course for contractors 0 42 Training Course for contractors-Sofala 0 29 Training Course for binec Sofala 0 4] Training Course for inspectors 0 42 Course in planning micro-projects (community particip.) 0 257 Course in s lannino economic projects 0 14 Course on toponomy in South Africa 0 2 Basic Courses on budget for project staff 20 0 Course on building maintenance 0 24 Participatory Rural Development (Community leaders) 60 0 Rural study for establishing land boundaries 0 ] 1 Course on building maintenance 0 25 Subtotal - Courses 120 487 Seminars Lecture on thematic cartography 0 1 Progress Seminar for the PRR in Zambezia 0 60 Introductory Seminar for the PRR in Sofala 0 40 Introductory Seminar for the PRR in Zamb&zia 0 40 Introductory Seminar for the PRR in Maputo 0 40 World Bank Procurement Seminar 0 5 Seminar to release study on the water sector 0 40 Seminar n micro-financ 0 1 Seminars with NGOs 40 0 Seminars on accounting for project staff 45 5 Seminars on management for project staff 60 0 Seminars on policy for government officials 45 0 Technical Seminars for project staff 60 0 Subtotal - Seminars 250 232 |Study visits Study visit to mnicro-projects in Gaza 0 2 Study visits to Namibia 0 3 Subtotal - Study visits 0 5 370 724 There were 29 participants in Sofala and 42 in Zambezia in the courses for contractors. They covered topics of special interest to contractors (contract law, legal aspects of contracts, logistics, organization of work, bidding, contracts, and other relevant subjects in this area), which improved the performance of the contractors and even the quality of the work. This experience demonstrated that the practice should be continued. There were 41 participants in Sofala and 42 in Zambezia in the courses for inspectors of civil works (which improved the quality of inspections and represented a substantial increase in the number of independent inspectors for rural community works). This is another element that should be repeated in future projects or activities of a similar nature. 36 The courses for inspectors and contractors involved more than 50 contractors and demonstrated that they are necessary and useful. The curriculum was developed on the basis of problem areas identified in inspectors' reports by project staff. These courses should continue to be offered in future projects, because of the results obtained in quality of work, speed of completion, careful attention to plans by the contractors, and attention to the contract clauses by contractors and inspectors. These courses were also useful for improving the working relationship between contractors and inspectors, both of whom assumed greater responsibility for contract compliance. In future projects it is important that this component integrated into the regular operational system of the public works sector, as a responsibility of this sector for providing this type of training for local contractors. Along with these courses there were two courses in Zambezia and Sofala on preventive maintenance for buildings, in which 40 people were trained. Participants were drawn from members of the communities and district employees in the public works sector. The courses demonstrated their value for sparking community initiatives to supplement the government's effort to maintain basic services for the population. Courses on planning with community participation. These courses were offered in 46 communities with a view to meeting the real needs of the communities in each area. The courses were taught with methods that permitted cornmunity participation in the identification, selection, and decision on community investments. At least 116 micro-projects were selected in this manner for financing by the Rural Services Rehabilitation Project. These activities helped people who felt ownership of the micro-projects. Successful planning of micro-projects with community participation through this project demonstrated that it is necessary to include this element in the national planning system so that iit is not an isolated and short-lived activity excluded from the national system of investment programming. The results of implementation of this component were considered satisfactory. Land Development Component National reconstruction and rehabilitation ol'agricultural services require maps showing land use and vegetation cover and an adequate policy to ensure security of land occupancy and tenure. Production systems and land ownership and use are essential factors in food production and environmental protection. This component was subdivided into two parts, namely Mapping and Soils. The soils sub-component was cancelled at the time of the first amendment of the project. Mapping The project called for: reviewing and updating maps on a scale of 1:50,000; determining the area of land occupied for each purpose; making detailed local maps to permit programming of activities at that level; creating a national capability to produce maps for various purposes in accordance with the needs of the various users, quickly and with satisfactory quality, through the purchase of appropriate equipment, training of personnel in this sector in the use of that equipment, and use of modern technologies for map production (processing of satellite imaging). The following specific activities were programmed: (i) update the national topographic map on a scale of 1:250,000; (ii) produce 500 maps showing land use at a scale of 1:250,000; (iii) produce statistics on land use and vegetation coverage; (iv) produce maps with 37 detailed data on eight districts at a scale of 1:50,000; (v) install the Geographic Information System (GIS) in the National Office of Geography and Cadastre (DINAGECA); (vi) install programs adapted to the system for the work to be accomplished; (vi) train specialists of DINAGECA to work with the equipment purchased, and operate and maintain the GIS; (vii) hold three seminars for approval of the work methodology and dissemination of the results of the first and second phases of the work; and (viii) train two university graduates in this area. The activities making up this component, with the exception of the training of two university graduates, were fully accomplished. All the work of the component was entrusted to a joint venture composed of CENACARTA (a Mozambican company) and IGN-France International (a French company). During the execution of the project, there was no problem finding financial resources for carrying out the various activities. Now that the project is finished, and considering the precarious financial condition, the sustainability of the effort is in doubt. The inputs and all activities related to map production are extremely expensive. Although the project left equipment and trained personnel, they cannot be used to their full capacity because of the government's limited resources, which do not permit doing high-volume jobs. The Cadastre objectives of rural services rehabilitation project were realistic because the various phases or stages were consistently finished on time. The rural services rehabilitation project was well designed. It is recommended that future projects pay more attention to aspects of training and technology transfer, and the planning and content of the training to be given. Along the same lines, the Provincial Geography and Cadastre Services (SPGCs) should have received special training so they could contribute more effectively to the refresher training that will be required in the future. This could reduce the costs incurred in sending technicians to headquarters to get the refresher training. We believe that dissemination of the database to the provinces, along with specific instructions for its management and use, is unrealistic because at the provincial level the necessary human and technical resources for this purpose are still lacking. An effort will be made to distribute the maps already produced so that they can be used. This aspect was not included in the performance contract for this component, so authorities in the sector are working to find funding for it and for gradually developing conditions for the provincial databases and dissemination of the work done. The financial resources obtained through the project and placed at the disposal of DINAGECA are making a significant contribution to the rapid and satisfactory meeting of the user demand for geographically referenced data. Maintenance is an integral part of the functional scheme of DfNAGECA as a governmental agency. Several alternatives are being explored for guaranteeing the continued functioning of the equipment. The results of the implementation of this component were considered satisfactory. Rural Water Supply Component This component envisioned increasing the service delivery area of rural water supply in Sofala and Zambezia, and improving the organization and management of the agencies responsible for water supply in the project area. Other objectives were to strengthen 38 community participation in the operation and maintenance of the water points by reinforcing the Community Education Program, and purchasing inputs needed for maintenance of hand pumps. Specific activities programmed were the following: (i) rehabilitate and construct a total of 120 water sources in Sofala and 180 in Zambezia; (ii) purchase of AFRIDEV brand hand pumps, which are easy to use and maintain, in replacement of the INDIA MARK 11 brand; (iii) make an inventory of water supply infrastructure in Sofala and Zambezia; (iv) undertake a study on the organization and management of area rural water supply companies (EPARs); and (v) increase output capacity (building three sanitary treatment plants for rural water in the districts of Upper Mol6cue, Chinde, and Buzi, and buying and delivering equipment for manual and mechanical well drilling), increase management capacity at water treatment plants (establishing systems for programming, monitoring, and coordinating of contract awarding, logistics, and community education), and improve the PEC through training for its administrators, extension agents, and operation and maintenance groups in methods of eliciting greater community involvement. In the initial design for the project, priority was given to strengthening the provincial treatment plants for rural water. This effort consisted of improving the output capacity of rura] water treatment plants in the provinces by construction of three new sanitary treatment plants and supplying various types of equipment, management training, and improvement of the mechanisms for increasing community involvement in water supply activities. The context of the war situation during the height of the design of thhe project, together with the lack of a strategy for the sector, had a pronounced effect on the shaping of these activities. After execution of the project was begun, unanticipated activities considered most beneficial for the sector were added. The most important new activity added was the study for institutional reorganization, which made it possible to draw up a policy and national strategy for water supply that is currently to be followed by the sector, and the inclusion of water systems to be funded by the rural services rehabilitation project. The activities can be summarized as follows: 39 Type Unit Projected Actual Purchase of well drilling equipment, manual and mechanical Sets 2 b Purchase of tractors and accessories N° 6 6 Purchase of two trucks NO 2 2 Purchase of hand water pumps NO 1,400 300 Construction of sanitary water treatment plant in the District of Upper Metros 133 0 Mol6cue Quadr. Construction of sanitary water treatment plant in the District of Buzi Metros 133 0 Quadr. Construction of sanitary water treatment plant in the District of Chinde Metros 133 0 Quadr. Planning prototype sanitary water treatment plant NO 0 1 Study of water supply system in Gurue NO 0 1 Study of water supply system in lie NO 0 1 Study of water supply system in Maganja da Costa NO 0 1 Study of water supply system in Pebane NO 0 1 Study for institutional reorganization of the water sector NO 0 1 Inventory of water points in Sofala and Zambezia NO 2 2 Inventory of water infrastructure in Sof./Zamb. NO 1 1 Construction and outfitting of water points NO 240 147 Rebuild and expand the water system of Gurue NO 0 1 A project for a prototype sanitary water treatment plant was developed with a view to expanding the geographic area of the provincial rural water treatment plants, as indicated in the project document. Since the project for expansion of the provincial rural water companies was abandoned in the implementation of the rural services rehabilitation project, the prototype can be used as soon as the future and legal status of the rural water companies is clarified. The National Water Office (DNA) is going to immediately commission a study that will assist in deciding the future of the current companies. The project bought 300 hand water pumps that were used in new and rehabilitated water points, as well as two trucks, six tractors, and two vehicles for the provincial water departments and the provincial rural water supply companies, to be used for transporting supplies and technicians for inspecting the water supply works. Absent a clear policy for the construction and rehabilitation of widely separated water points, the purchase of hand water pumps was somewhat limited until the strategy to be followed is clarified. The inventory of water sources in Sofala and Zambezia provinces was completed. The data are being used for programming the water supply. This study also produced software for the database on water sources that is being used by the water departments in Zambezia. In Sofala the database was not used for lack of adequate follow-up. A study was done for the institutional restructuring of the water sector. Its recommendations are to be implemented and have guided the activities of the National Water Office. This study was done instead of the study on management and organization of provincial rural water treatment systems, because of the need to define the strategy for the water sector before that study is begun. The water sector will use the recommendations of the study that was completed. 40 Studies were done on small water supply systems (Gurue, lie, Maganja da Costa, and Pebane). Among the water supply systems studied, thie one for Gurue was rehabilitated and enlarged. This was carried out and completely finished. As for the management of this system, the National Water Office has already begun a study that will help to identify management options for similar situations. The results of the implementation of this component were considered satisfactory. Soils and Resettlement of the Population This sub-component and component were cancelled. LESSONS LEARNED Planning of micro-projects with community involvement The process of community participation in the process of investment planning should be continued by being institutionalized. This would ensure its permanency because it would become part of the national investment programming system, and would avoid the interruption in the process that occurs when officials are transfer-ed out of the district or province. Project timeframes Whenever possible, projects that involve cornmunity participation in their execution, maintenance, and operation should have long time fiames, and be free of disbursement pressures, which generally tend to interfere with the best organization of such processes as the development of procedures and the necessary involvement of the appropriate agencies. The question of community participation is a long-term process that cannot be expected to proceed at the same pace normally anticipated for disbursements. Flexibility in the execution of projects of this kind, especially those that involve multi-sector coordination, should be taken into account in project design so that positive results likely to improve the implementation process can be incorporated. Institutional development and training It was unrealistic to expect this project to achieve much in the way of institutional development, which depends on institutional development at the central level. It is essential to have clear rules and procedures for the execution of works at the provincial and district levels, with well-delineated lines of responsibility for these levels in the various sectors. This would facilitate clarification of project activities at every level, bearing in mind that decentralization goes beyond the physical aspects of investment decisions to include questions of financing. It is to be hoped that future projects of this sort are preceded by studies that reconcile provincial and regional concerns and those of the responsible central agencies with project objectives. Inspection and civil construction companies benefited more from the training effort, since their problems had been better identified. 41 Support for provincial and district public agencies Future rural development projects should devote more attention to providing these institutions with support through the central sector agencies, particularly in the definition of standards, procedures, and training. Training Training and orientation for contractors and inspectors in their respective fields and others such as contractor law, legal aspects of contracts, logistics, organization of work, competitive bidding, contracts and other relevant subjects, should be a permanent element because it proved to be one of the best ways to improve the performance of contractors and inspectors, and to raise quality and speed in the execution of works. It is a practice that should be continued in future projects of this type. Courses on preventive maintenance of buildings The courses proved themselves to be necessary in eliciting community initiatives to supplement government efforts to maintain basic services for the population, and should be considered in future projects. Mapping and land use The activities of this component were all entrusted to a specialized company, IGN- FRANCE INTERNATIONAL/CENACARTA, and were completed as a whole. One job that remained unfinished is to be continued with the training of the provincial geography and mapping services, which will be responsible for updating the mapping in each of the provinces. These units would carry on the work continuously in each province and transmit the results to headquarters in Maputo for periodic updating of maps. The sector will seek funding to permit dissemination of the results of the mapping that was done, and to train provincial technicians in the use of the results obtained. Rural water supply Except in emergency situations, water supply programs should respond to actual demand and be carried out by private management or using other water supply system management models. Studies on various altemative management systems are going to be carried out by the National Water Office. One of the systems to be considered is that of Gurue, which was rehabilitated and expanded with funding from the Rural Services Rehabilitation Project. 42 IBRD 24277 CONGO 9UAL~~~~~DI 3g4 36-W40 P ts ZAIRE TOOZANL& > 34 ( \ 36 3&ATZI w z\ X NIASSA , V {CSr MA LA M _.fPua] . ZAMBIA ~ ~ ~ ~ ~ TAZAI M..O C.h.. B... D- ~ ~ ~ ~ Mcoo HARARE ~~~~~~~~~~~MOZAMBIQUE c RURAL REHIABILITATION PROJECT Z A M B I A ti < 1° I3^J~~~~~NOE Prol^ect Supported Distridts In Zambezia- \ CHiNDE . 2 [ > uIINDE (+) F~~~~~~~~~~~~~~~IROVINCE~t<>AL / O~~~~~~~~~~~~~~~~ ADMINEMWE POSTS / ~~~~~~~~~~~~~~~~~~~DISTIUC BOUNESP X - RB i~~~~~~~~~~~~~~~NION 30UNDARES / - NXERNA~~~~~~~~~~MOWL BNOANME on This map dos beeno,n ahed par of Ttioe Worl 8ber0knon9pc arty judgn" on. ttrw er wdrst or atuetny CLOMMlR S suhbourldafiesW ptxtfAekl FEBRUAIRY 1 993 IBRD 24279 ON -7GO eWMO*A Xe>- NAS. MOZAMBIQUE RURAL REHABILITATION PROJECT ram a Project Supported Districts In Sofala \ k < g> > CHEMBA *) - PROJECTAREAS a SOAFuRTIH Ct Ocs \ CHEMBAX 2 i PROVINCE CAPITAL ________\___Mulin__ _ a /S 1 * DISTRICT CAPITALS This map hos been prepared Canxixe 1 ADMfNISTRATIVE POSTS by The World Bank s staff 0 - DISTRICT BOUNDARIES exclusively for the convenience of readers and is for the C PROVINCE BOUNDARIES internal use of The World Bank CAIGUA A - ITRAINLBUDRE Group.The denominations used MARINGUE CAL R and the boundaries shown I B A on this map do not imply, on the part of The World Bank Goup, RNU any judgment on the legal status ot any territory or any /ty.. endorsement or acceptance of upang such boundaries. \ g \ Su~bu4 ,Y anhm nga \t ,Nh modzi \\ MARRO MEU\ U A qyi 7 ( Vunduzi MARRO 'INHAMINGA 0 50005GORONGOSA 250 ORONGOSA f \ < > Ivk~~~NUANZ J '-^.A-9Er r ~FAMUANZA 199 KWaANDA t> NHAMATANsDA 9 g > BtI~~~~sZI E\,A 1 A . I1 N D I A N 13 G nda/ BUZ I CHIBABAVA St>folo ,. t HIBABAVA .- > < O~~~~~~~~~~~~~~~~~~~ C E A N 1i 0 / Dvnhe 4 . b un~~~~~Mxugue / 0 i\ A ~~~MACHANGA MACHANGA % ,( it9~~HAMBANE \ 0 50 1 00 1 50 200> 250 KILOMETERS Approximate Scale