Algeria - Cadastre Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 22298 IMPLEMENTATION COMPLETION REPORT (CPL-34880; SCL-3488A; SCPD-3488S) ONA LOAN IN THE AMOUNT OF US$33.0 MILLION TO DEMOCRATIC REPUBLIC ALGERIA FOR A CADASTRE PROJECT June 6, 2001 Financial Sector, Private Sector and Infrastructure Middle East and North Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (Exchange Rate Effective June 2001) Currency Unit = Algerian Dinar DAI.00 = US$ 0.012 US$ 1.00 = 79.6DA FISCAL YEAR January 1 to December 31 ABBREVIATIONS AND ACRONYMS ANC Agence Nationale du Cadastre (National Cadastre Agency) APC : Assemblee Populaire Communale (Municipality) BA Banque d'Alg6rie (Central Bank) CF Conservation Fonciere (Land Registry Office) CNTS Centre National des Techniques Spatiales (National Center for Land and Space Technology) DAS Domaine Agricole Socialiste (State Farm) DGDN : Direction Generale du Domain National (General Directorate of Lands) EAC : Exploitation Agricole Collective (Privatized Collective Farm) EAI : Exploitation Agricole Individulle (Privatized Individual Farm) EPA Etablissement Public Administratif (Autonomous Public Agency) GoA : Gouvernement d'Alg6rie (Government of Algeria) INC Institut National de Cartographie (National Mapping Institute) LIS Land Information System MoA : Ministere de I'Agriculture (Ministry of Agriculture) MoE Ministere de l'Economie (Ministry of the Economy) MoF : Ministere des Finances (Ministry of Finance) MoH Ministere de l'Habitat (Ministry of Housing) MoJ : Ministere de la Justice (Ministry of Justice) Vice President: Jean-Louis Sarbib Country Manager/Director: Christian Delvoie Sector Manager/Director: Sonia Hammam/Emmanuel Forestier Task Team Leader/Task Manager: Douglas I. Graham/Mayumi Kato  FOR OFFICIAL USE ONLY ALGERIA ALG-CADASTRE CONTENTS Page No. 1. 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 5 5. Major Factors Affecting Implementation and Outcome 8 6. Sustainability 9 7. Bank and Borrower Performance 9 8. Lessons Learned 11 9. Partner Comments 13 10. Additional Information 21 Annex 1. Key Perfornance Indicators/Log Frame Matrix 22 Annex 2. Project Costs and Financing 23 Annex 3. Economic Costs and Benefits 25 Annex 4. Bank Inputs 26 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 28 Annex 6. Ratings of Bank and Borrower Performance 29 Annex 7. List of Supporting Documents 30 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.   Project ID: P004964 Project Name: ALG-CADASTRE Team Leader: Douglas I. Graham TL Unit: MNSID ICR Type: Core ICR Report Date: June 12, 2001 1. Project Data Name: ALG-CADASTRE L/C/TF Number: CPL-34880; SCL-3488A; SCPD-3488S Country/Department: ALGERIA Region: Middle East and North Africa Region Sector/subsector: UY - Other Urban Development KEY DATES Original Revised/Actual PCD: 06/19/1990 Effective: 02/24/1993 Appraisal: 10/11/1991 MTR: 10/04/1996 Approval: 06/11/1992 Closing: 12/31/1997 12/31/2000 Borrower/Implementing Agency: GOVT. OF ALGERIA/NAT'L CADASTRE AGENCY Other Partners: GENERAL DIRECTORATE OF L ANDS STAFF Current At Appraisal Vice President: Jean-Louis Sarbib Caio Koch-Weser Country Manager: Christian Delvoie Pieter P. Bottelier Sector Manager: Sonia Hammam Amir Al-Khafaji Team Leader at ICR: Douglas I. Graham Vincent Lacour ICR Primary Author: Mayumi Kato 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: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time:  3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: 3.1.1. Original Objective. The objectives of the Project was to assist the Borrower in establishing a land registration system (cadastre), which system includes a general cadastre and a land registry, throughout its territory, by supporting the first four year period of the General Cadastre Establishment Program (GCEP). In particular, the Project would assist the Borrower in; (i) developing the appropriate legal and institutional framework for the land registry; (ii) gradually increasing activities related to the land registry during the first four year period of the GCEP; and (iii) organizing computerized and other information systems, based on the Cadastre, for taxation and other land related uses. The Project, through accelerating the development of the cadastre system in Algeria, was envisaged to contribute to the development of the agriculture and housing sectors through stimulating the development of both agricultural and urban land markets. 3.1.2. Background. In the early 1990s, as part of the ongoing economic reform, the Govemment identified Housing and Agriculture as the two most crucial sectors for accelerating economic growth. Improving agricultural productivity was a priority for the Algerian economy mainly due to the following three factors: (a) rapid population growth; (b) heavy dependence on food imports; and, (c) modest share of usable land in the national territory. The main element of the agriculture development strategy was to privatize state farms to a number of individual farmers to increase agricultural productivity. As part of the economic reform process, the GoA decided in 1987 to dismantle the state farms which represented 40 percent of the domestic agricultural output and to give usufruct rights to about 27,000 private groups of farmers. In the economic reform program, the GoA set a strategy to deepen this land privatization process and enhance private sector activities in agriculture. 3.1.3. In the housing sector, the Government's strategy mainly focused on shifts from a supply-driven, public sector dominated strategy to a private sector driven and market-oriented sector in order to address more effectively the chronic shortage of urban housing. In 1990, the new Land Policy Law was enacted to enable free land transactions of urban land. Prior to this law, transaction of land among private parties was legally prohibited since municipalities (APC-Assembi6e Populaire Communale) were monopolist suppliers of urban land. APCs were unable to respond to the growing needs for developed land due to lack of incentives and lack of management capacity. Pricing of urban land was also politically determined and did not reflect the true market value; consequently, there was a constant excess demand for urban land, and sustainable urban growth was always hindered. 3.1.4. In order to further develop the agriculture and housing sectors, clear registration of land titles was considered as the most critical step. In agriculture, clearly registered titles would facilitate land transactions and stimulate farmers to undertake long-term planning of agricultural investments, and enable possible consolidation of land to improve agricultural productivity through private transactions. In housing, clear titles would stimulate the rapid development of urban land and housing market through easier transactions and development of mortgages, and would lead to the economic use of scarce urban land. However, in the early 1990's, reportedly only 5% of private rural and urban owners had clear land titles. Thus, development of a reliable cadastre system to register and deliver land titles was considered as a most urgent task. 3.1.5. Actually, the effort to develop a cadastre system started in early 1863 in Algeria but was limited to a small part of the territory. The system became operational with a comprehensive legal framework for land registration in 1975. This system mainly targeted the northern region of the country covering approximately 110,000 km2 where the majority of the population lived. However, the development of the cadastre system has been slow due to a combination of several factors such as: lack of funds and expertise, and duplicated and inefficient institutional settings. In the late 1980s, the system covered only 5 percent of rural areas and less than I percent of urban land. 3.1.6. Among all the above factors, reform of the institutional setting was considered most important. In the early 1990s, two institutions were involved in developing the cadastre system: The National Cadastre Agency (Agence Nationale du Cadastre, ANC) and Land Registry Office (Conservation Fonciere, CF). Both ANC and CF were established under DGDN (Direction General du Domain National: General Directorate of Lands) of Ministry of Finances. Duplication of tasks and lack of proper coordination between the two institutions were often - 2 -  observed. ANC had responsibility in mapping and cadastre establishment, while DGDN/CF was in charge of parcel adjudication and land register maintenance. Within this system, land rights were investigated twice, first by ANC and again by the CF. Under this arrangement, not only development of the system was slow, but also the quality of the system was not reliable, due to: (i) lack of clear cadastre establishment procedures, and (ii) unreliable conveyancing and cadastre maintenance. There was an urgent need to redefine mandates of these two institutions, establish clear procedures to eliminate duplications, and provide adequate means to increase efficiency. 3.1.7. In response, the Government developed a strategy in collaboration with the World Bank in the form of a new General Cadastre Establishment Program (GCEP), aiming at restructuring institutional setting for developing the cadastre system and completing the title registration by year 2008 (15 years from 1993). The Cadastre Project was designed to provide the Govermnent with targeted support to start implementing the GCEP. The Project was appraised in November 1991. 3.2 Revised Objective: Not applicable. 3.3 Original Components: 3.3.1. The project comprised the following three components; Component 1. Cadastre Mapping and Parcel Adjudication (US$29.6 million, Revised amount: US$24.7 million). This component aimed at supporting the production of cadastral maps and the adjudication of parcels for 3 million ha of rural land and 36,000 ha of urban land. It also included the testing of various urban mapping techniques and introduction of computerized parcel registration and map production. In particular, following activities were planned under this component: (a) launch of an aerial photography program and improving productivity in INC (Institut National de Cartographie: National Mapping Institute); (b) pilot surveys of four urban zones and development of stereoplotting systems for urban cadastral mapping; (c) improving productivity of ANC ground survey teams and CF offices; and (d) promoting the use of computers for land registration and map drawing. Component 2. Institutional Support to ANC. (US$2.3 million, Revised amount: US$2.7 million). This component mainly financed technical assistance and equipment to upgrade the expertise of ANC and to increase its operational efficiency. In particular, this component financed the following activities: (a) preparation of a restructuring plan to show ANC's new organizational structure; and (b) development of a MIS (management information system), including financial management, and monitoring aspects. These activities were expected to increase efficiency by simplifying the existing procedures for cadastre registration and facilitating day-to-day monitoring of ANC 's performance. Component 3. Technical Assistance and Studies to other Governmental Agencies (US$1.1 million, Revised amount: US$0.5 million). This component aimed at supporting other agencies than ANC which were involved in issues related to cadastre and land management systems. The following activities were planned: (a) training activities for DGDN/CF; (b) technical assistance to the Ministry of Justice for the organization of land courts in all wilayaete to strengthen legal aspects of the titling procedure; and (c) preparation of a study on land tenure by the Ministry of Agriculture aimed at identifying constraints to agricultural development that emanate from issues of rights over land and at developing the policy framework to address these constraints. 3.4 Revised Components: 3.4.1. The project experienced three extensions and 2 cancellations as follows; 3 extensions: 1) December 1997 (the original closing date) extended to June 1999 2) June 1999 extended to June 2000 3) June 2000 until December 2000 -3 -  2 cancellations: 1) June 1999, US$2.27 million 2) June 2000, US$2.4 million 3.4.2. The first extension in 1997 was agreed when GTZ started a project which provided additional technical assistance in support of the Bank's project. Through this extension, the project aimed at focusing more on the DGDN/CF after making initial achievements in the ANC. The second extension in June 1999 was requested because of the implementation delays caused by the lack of counterpart financing. Counterpart funds were made available in two trenches (October 1999 and March 2000) followed by the second extension. Ministry of Agriculture launched a procurement of a study on agricultural land in October of that year. However, the study only started in March 2000 and thus the third extension was requested to allow completion of the study. 3.5 Quality at Entry: 3.5.1. Quality at Entry was marginally satisfactory. The project rightly aimed at accelerating land registration through various means, such as strengthening institutions, streamlining the procedures, and modernizing the system. The technical approach proposed under the project was also appropriate, considering the capacity of the respective institutions. 3.5.2. In retrospect, the project impact would have been more substantive if more attention had been paid to improve the capacity of the DGDN/CF. At appraisal, it was rightfully judged that the consolidation of the ANC and the DGDN/CF was politically difficult, and that due attention has been paid to simplify the procedures to register infornation into the cadastre system through eliminating duplication. In this context, ANC and DGDN/CF were to continue as the two major public institutions involved in the cadastre system; ANC is responsible for land information aspects (e.g. topographic survey), while DGDN/CF are responsible for legal aspects (title and registration). However, the project's support was targeted largely to the ANC (e.g. Component 2), and inadequate attention was paid to strengthening the capacity of DGDN/CF. During project implementation, it was noted that limited capacity of DGDN/CF had indeed become a constraint for timely registration. While several supervision missions recognized this issue and a training program for DGDN/CF was developed within the project on an ad-hoc basis, it will take some time to manifest impacts of such technical assistance. 3.5.3. In addition, the project envisaged a technical assistance program for the Ministry of Agriculture and Ministry of Justice without defining clearly their roles, although both were regarded as main partners in the cadastre process (Component 3). This rather complex institutional arrangement required a large amount of coordination among various public institutions, and has contributed to the delay in project implementation. As was the case for ANC and DGDN/CF, implementation of the technical assistance for these two ministries faced difficulties arising from lack of counterpart funds. 3.5.4. Furthermore, quantitative targets of the project on cadastre map production and adjudication were too ambitious. The Staff Appraisal Report (SAR) targeted to cover 3,000,000 ha in rural areas and 36,000 ha in urban areas by 1997. The project focused mainly on improving mapping techniques in ANC. However, the quantitative target was set on the overall process, including parcel adjudication. While the target was supposed to be developed based on pilot exercises in rural and urban areas using various mapping techniques, the target figure was simply over-optimistic, considering the capacity of the institutions involved and specific characteristics of each wilaya (district) covered under the project. While ANC progress was substantial, DGDN/CF, in charge of titling, was eventually recognized as not having sufficient capacity to adapt to this target. A program aimed at filling the gap between ANC and DGDN/CF only started in 1998, but the implementation duration was too short to provide results. The project therefore required three extensions amount to a total of 3 years to reach the target set for the original project duration of 4 years. At the time of this ICR, 96 percent of the SAR target was achieved for rural areas and 144 percent for urban areas. This indicates that the project, when it initially determined the quantitative targets, underestimated the time and capacity needed to achieve them. The project has led ANC and DGDN/CF to considerably improve their efficiency. -4 -  4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective. 4.1.1. As stated in Section 3.1, the project objectives are to: (a) strengthen the institutions that implement the cadastre; (b) support a gradual increase of cadastre-related activities (e.g. topography, mapping) over a four-year period; and (c) develop information systems based on cadastre for taxation and other land-related purposes. The project largely met the two first project objectives and met the third objective only partially. 4.1.2. The project was instrumental in collecting topographical information and developing mapping, which are essential for the cadastre system (Objective (b) above). Institutional strengthening for ANC was also effective in increasing the productivity of mapping (Objective (a) above). In particular, at the time of project completion, approximately 2.9 million hectares, or more than 95 percent of the target (3 million hectares) had been mapped for rural areas, and 52,000 hectares , or 85 percent of the target (60,000 hectares) for urban areas. Under the project, the coverage of the cadastre system has more than doubled (from 196,000 hectares in 1993 to 400,000 hectares per year at project completion) for rural areas. For urban areas, the project was instrumental in initiating the cadastre system (the annual area registered has increased from less than 500 hectares to 15,400 hectares per year. This increase is due to the fact that productivity has increased dramatically. Productivity measured by hectares mapped per team per year has increased from 20 to 30 hectares for rural areas, and from 23.4 to 158 hectares for urban areas due mainly to technical assistance, streamlined procedures, and upgraded hardware. 4.1.3. To a certain extent, the project has also contributed to the institutional development of DGDN/CF. Although operational results remain modest by project completion, DGDN/CF management has become nevertheless fully aware of the need for pursuing the reform program and more resources to implement it. To continue these efforts, it is being suggested that DGDN/CF request new financing through the Housing Finance Project under preparation with the Bank. Overall, the training program conducted under the project strengthened morale among key staff. It has nurtured motivation among them in improving the quality of work and reliability of the information registered into the system. However, as for DGDN/CF, the training program intended for them was carried out at a relatively late stage of implementation, and the impact of the training was relatively minor at project completion. In addition, the training program has simultaneously revealed other constraints such as inadequate intemal communications and poor coordination among different institutions, due to poor initial institutional arrangement. Under such circumstances, it was impossible for the trained staff to fully utilize their learning in practice before completion. 4.1.4. Objective (c) above has been achieved partially at completion. While the delivery of titles would be critical to introduce property taxes, at project completion, the Government has not yet introduced the concept of property tax systems. Titles were prepared for 464,796 lots. Out of these, the project was able to register 149,605 titles by project completion, and more than 40 percent of the titles (64,371 titles) have been delivered to the respective owners. The project also assisted in preparing a land information system but did not go further than the pilot phase (objective (c) above). The selected system is an object oriented geographical information and archiving system. The software is now being installed, after having been tested at a pilot phase. As initially planned, the cadastre will become the basis of the taxation system only when the titling is fully registered and the computerized tools are fully installed. In addition, through closer collaboration between ANC and DGDN/CF, accuracy of titling information obtained in the field has improved. This has contributed to more efficacy in preparing titles. It should also be noted that the project has contributed to the development of the property market. The number of titles prepared is expected to increase with a rate of 237,600 titles per year, and impacts related to this objective would be more visible in the very near future. 4.1.5. In conclusion, the project was instrumental in supporting the Government to take initial steps in developing a cadastre system. It has certainly increased Government's capacity to carry out the General Cadastre Establishment Program (GCEP). The project contributed to the establishment of technical methodologies and provided critically needed equipment and developed human capital. After the completion of the project, it is expected that the Government will continue to implement the GCEP. The current speed of the implementation is following the benchmark set out in the 15 year model described in the study prepared under the project. It should be noted that through implementing the project, the benefits of merging ANC and DGDN/CF field teams. In -5-  addition, the Government became aware of the need to merge the DGDN/CF and the ANC to accelerate the development of a cadastre system, and has started internal discussions to proceed with the possible merger. 4.1.6. Among the remaining challenges are: (a) development of fiscal cadastre systems, which would be the basis for the property taxation systems; and (b) introduction of a cost recovery mechanism. Development of a fiscal cadastre system, together with a property valuation system, was not included within the project scope. But its future importance as the next step is obvious. A cost recovery mechanism, including quality control and cost accounting system, shall eventually be introduced to make the cadastre system fiscally sustainable. However, the Government decided that fees would be charged to land owners only for transactions and not for initial cadastre setting. In addition, technical standards of some equipment procured through the project have become outdated over time. They require maintenance and updating to sustain current productivity level. The final supervision mission discussed these issues with the GTZ, which agreed to provide follow-up technical assistance. It is also expected that the proposed Housing Finance Project to be financed by the Bank would follow up on these issues. 4.2 Outputs by components: 4.2.1. Component 1. Cadastre Mapping and Parcel Adjudication (Estimated Cost: US$ 65.3 million - Actual Cost: US $ 48 million). The following activities have been carried out under this component: Improving branch offices (antennes locales) of the ANC. The project design included the improvement of 28 branch offices of the ANC covering the northern part of the country. GoA decided to finance this component fully in local currency, and requested an amendment to transfer the equivalent amount to component 2. A total of 16 sites have been completed and works on procurement of works for the remaining 12 sites have started in accordance with the project objectives. Including the additional 9 local offices being housed in the local offices of the Ministry of Finance, a total of 36 local offices are now established for the cadastre and titling services. Mapping works have dramatically improved in speed and quality, with improved working environment and updated equipment. In the past, the staff of the ANC had been forced to place their work desks in the corridors and all mapping works were han('led manually. In addition to all branch offices, five regional offices of ANC and 182 local offices of CF and the Headquarter of ANC were also computerized. Aerial Photography. Undrer this component, a specialized airplane and photographic equipment for the INC ( Institut National de Cartographie: National Mapping Institute) have been purchased. The INC has agreed to provide aerial photographs for the ANC. This arrangement has accelerated the overall mapping process for the ANC, particularly in urban areas, as aerial photography turned out to be 7 times more productive than level topography. The dramatic impacts on production stemming from aerial photography resulted in the decision to equip a second plane, owned by INC, with high quality digital cameras. Quality control system and cost accounting system for the ANC. Under the component, technical assistance was financed to prepare a design of these two systems. This study covered the following aspects: (i) identification of appropriate indicators to monitor quality and productivity; (ii) experimentation of cost accounting modules in coordination with GTZ technical team; and (iii) the formation of a consultative process in defining yearly objectives of mapping production. However, the development of the designed systems was not completed, as it was considered to be premature. Within the limited resources and human capacity available, it was decided that efforts should be concentrated on increasing the coverage of mapping and increasing the number of delivered titles at the very initial process in developing the cadastre system. Nevertheless, some results were achieved : a cost accounting system has been tested on one wilaya, and the consultative process of defining quantitative objectives started in 2000. It was considered a success among managers and improved communication between management layers. As stated in Section 4.1, GTZ has agreed on following up on developing quality control and cost accounting systems after completion of the project, and it is expected that the study financed under the project would form a basis for GTZ's activities. 4.2.2. Component 2. Institutional Support for the ANC (Estimated Cost: US$ 5.8 million - Actual Cost: US $0.005 million). The component mainly financed training and technical assistance. Under this component, comprehensive training courses covering administrative, technical, and managerial aspects were offered to the - 6 -  majority of the ANC staff: the project, in coordination with GTZ, financed a comprehensive training program for high level staff to be held in France and Germany. ANC created a training unit to assist all operational staff in adapting to new technology. During project implementation, high level DGDN/CF staff was also included in the training under this component. At a later stage, DGDN/CF also established a unit in charge of training programs (Cellule de Formation et de Communication), in order to develop a long-term training strategy targeting all staff. Management and Organization Study. A study on management and organization of ANC was conducted in 1993. The study defined and proposed a new organizational structure and information network. According to the proposed information network, 6 pilot stations were set up in 1996 and 100 computers were procured. 4.2.3. Component 3. Technical Assistance and Studies of other Governmental Agencies (Estimated Cost: US$ 1.4 million - Actual cost: US$ 2.9 million). Under this component, technical assistance was provided to the Ministries of Justice and Agriculture. Ministry of Justice. The technical assistance was provided with the following two objectives; (a) to increase efficiency of lawyers in the judicial phase of the titling process; and (b) to facilitate the establishment of a judicial system related to land issues. In particular, the technical assistance was provided to establish land courts, train lawyers, and to acquire documentation for the creation of a judicial database. At completion, 48 land courts were established covering all wilayete by 1998. In addition, training on land conflict issues was provided to facilitate the initial operation of the land courts. However, the training was not completed before the loan closing due to the lack of counterpart financing. The establishment of a legal database was completed and the land courts were equipped with documentation. Ministry of Agriculture. The technical assistance helped to prepare a study on agricultural land tenure, which posed politically sensitive and technically complex problems. These problems involve more than just land privatization, which affects only about 35 percent of arable land and less than 10 percent of agricultural land, including the parcours and steppes used for grazing. The study intended to address the following main aspects: (i) sale and lease of public lands distributed to collective and individual farms (Collective farms have been effectively divided up into private farms); (ii) completion or restitution through settlement of disputes over land tenure and cases of prior restitution claims; (iii) development of land markets and associated institutions; (iv) development of a regulatory framework on traditional communal rights over the steppes and marginal lands; (v) titling on traditionally private lands; and, (vi) completion of cadastre activities. The study developed a set of recommendations and a priority action plan. The final analysis of the Study proposed that a long-term approach and commitment are required to deal with the above issues on agricultural land tenure. In parallel with the Study, legislation related to agricultural land tenure is being discussed in the Parliament, in which the Bank played a consultative role. Once this legislation is passed, need for related regulations and a program to assist market organizations may arise. Concessions over land also needs to be reviewed and assessed. In this context, the study has stimulated discussion at technical and policy levels of the Government. The study has been instrumental in channeling the focus on agricultural land tenure discussions towards technical, administrative and policy issues and providing a sound basis for future activities in this domain. 4.3 Net Present Value/Economic rate of return: No ERR/NPV calculation was carried out at appraisal. 4.4 Financial rate of return: Not applicable 4.5 Institutional development impact: 4.5.1. Refer Section 4.1.5. The Project has made a substantial institutional development impact on two executing agencies (ANC and DGDN/CF). Although these two agencies have yet to proceed with further reorganization and improvement in managing capacities, the Project has set the ground for institutional development. -7-  5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: 5.1.1. There were two major factors beyond the control of the Government which significantly delayed project implementation: (a) deterioration of the security situation between 1994 and 1996; and (b) the decline in oil price between 1997 and 1998; 5.1.2. The serious security problem, particularly in the northem rural area, made it almost impossible for Govemment staff to carry out topographic surveys and related data collection in the field. In particular, the implementation of the project was delayed in regions surrounding the Algiers area and in the eastem part of the country, such as Wilaya Tizi Ouzou and Wilaya Msila. 5.1.3. The decline in the oil price during 1997 and 1998 seriously affected project implementation, too. As the state budget is always heavily dependent on revenue from oil sales, counterpart funds in local currency were not provided on a timely basis between 1997 and 1998, when implementation of the project was at its peak. While overall expenditure for the project was approved, under the temporary fiscal measures called AP (Autorisation de programme) and CP (Credit de Paiement), disbursement of counterpart funds scheduled in 1998 were postponed by more than ten months. This slowed down the project progress in 1998 and 1999. 5.1.4. In addition to these two factors, unreliable telecommunication made it very difficult for the central offices of the ANC and the DGDN/CF to monitor progress in local offices and to take due actions in a timely manner. 5.2 Factors generally subject to government control: 5.2.1. The Govermment was slow in simplifying the procedures and institutional arrangement. Among the most serious factors contributing to the delay in project implementation was a delay in consolidating the procedures between the two institutions: DGDN/CF and ANC. Consolidation of procedures aimed at simplifying the institutional arrangement for the cadastre registration and accelerating the process. In addition, centralized and traditional management of these two agencies resulted in unnecessarily time-consuming procedures for implementation and lack of initiatives by local staff. The impact of this centralization was even exacerbated in conjunction with unreliable telecommunication (refer Section 5.1.4.). 5.2.2. In addition, there are several bureaucratic and unnecessary procedures which have also delayed project implementation; for example, the approval of the municipalities (APC: Assemble Poplulaire Communal) was required to initiate the procedures for cadastre mapping; however, communication between the ANC and the APCs was not always smooth. Often it took more than 5 weeks for the APCs to process such requests, as these APCs did not fully understand the needs for cadastre registration. 5.2.3. Because of a lack of commitment from the Government on counterpart resources over the project implementation period, the annual budget was dependent on extemal factors, such as oil prices, as described above. Therefore, the implementing agency had to compromise between a multi-year activity program and unreliable annual resources. 5.2.4. Disbursement procedures involving the BA ( Bank of Algeria) were also unnecessarily complicated and contributed to the delay in the project. 5.3 Factors generally subject to implementing agency control: 5.3.1. In principle, the performance of ANC and DGDN/CF was satisfactory (refer Section 7.6). Both of these agencies assigned competent staff dedicated to the implementation of the project. Project administration, including disbursement and procurement, was satisfactory in general. Annual project audits revealed no major issues. 5.4 Costs andfinancing: 5.4.1. Changes in project costs occurred between project appraisal and project completion. Total project costs increased from $96 million to $120.4 million with reduced Bank financing (from $33 million at appraisal to $27.9 -8 -  million at loan closing). Due to the difficulty in obtaining counterpart finance in local currency, as mentioned above, the parts to be financed by the Bank was downsized. To resolve this difficulty, GTZ provided a grant of $82.4 million to cover mainly civil works (Civil works which were to have been financed by the Bank at appraisal, were financed by Government and GTZ.) , recurrent costs and VATs. 6. Sustainability 6.1 Rationale for sustainability rating: 6.1.1. Sustainability of project impact is likely. As stated in Section 4.1, the project contributed to initial implementation of the GCEP through institutional support for the two key institutions in the cadastre sector: the ANC and the DGDN/CF. The methodologies and systems introduced under the project will continue to be used by these two institutions to implement the GCEP. However, recurring costs for the maintenance and upgrading of the equipment acquired was not foreseen and this needs to be resolved to sustain the project impacts. On procedures, the Government has by now taken all steps to eliminate the duplication between ANC and DGDN/CF which had been the most important institutional constraint to accelerate the implementation of the GCEP. In addition to the merger of operational teams (brigade mixte), these two agencies have also set out a joint process of planning training programs through the project. As for cost efficiency, ANC intends to outsource topographical works to private surveyors. After a long transitional phase spent in facilitating the creation of a surveyors' association and organizing training sessions, ANC launched the first series of bidding in late 2000, with the expectation that this could reduce the cost of registration and thus speed up the overall process. 6.1.2. While the importance of cost recovery from users was recognized at appraisal as being necessary to establish sustainable cadastre system, no major actions has been taken in this end under the project. In this context, it cannot be concluded that the sustainability of the system is fully ensured. However, this decision may well be justified, as at the very initial stage of developing the cadastre system, it was more important to increase the geographical coverage of the area and increase the reliability of the registered information. Indeed, the Bank supervision mission in 1995 supported this decision not to require beneficiaries to pay fees for the issuance of title in order to accelerate the progress, since it recognized that beneficiaries might not be willing to participate in the project, as registering in the cadastre system would imply future taxation on their property. In retrospect, this was a wise decision, as the ANC and DGDN/CF were able to accelerate the title establishment. 6.1.3. Both the ANC and the DGDN/CF are aware of the importance of cost recovery, and requested GTZ, the co-financier of the project, to provide continued support, mainly technical assistance, to prepare a plan to introduce a cost recovery mechanism. In response, GTZ is planning to first evaluate the economic impacts of the cadastre systems, and then prepare a concrete plan to increase quality of the information registered in the system and then, in the longer term, develop a fiscal cadastre system. 6.2 Transition arrangement to regular operations: 6.2.1. In addition to the continued support of the GTZ (refer Section 6.1), the Bank would provide follow-up technical assistance mainly targeted to DGDN, to support a comprehensive reorganization of the directorate. This request, considered as the next priority by the Government, is now being considered for possible financing under the proposed Housing Finance Project. 7. Bank and Borrower Performance Bank 7.1 Lending: Lending: Unsatisfactory 7.1.1. The Bank's performance during project preparation was considered marginally satisfactory. Whereas the overall project objective to provide institutional technical assistance was indeed appropriate at the very initial stage in developing the cadastre system, the project also contained activities, such as studies for the Ministry of Agriculture, which were premature in retrospect. In light of the limited capacity of the implementing agency, this blurred their focus, and caused delay in project implementation. -9-  7.1.2. In the meantime, the project preparation team did not elaborate on the optimal institutional set-up for the cadastre system. The initial project design did not also pay due attention to the importance of strengthening the DGDN/CF. In retrospect, the impacts of the project, mainly delivery of titles, would have been much more substantial, if the institutional arrangement had been completed at the start of the project, or more substantial institutional support for the DGDN/CF had been considered at appraisal so that the project could have focused on identification of the technical methodologies and development of the registration procedures. 7.1.3. The quantitative targets set during project preparation were also ambitious. A project preparation study carried out during project preparation set out two options: 10 year and 15 year to complete the GCEP. For rural areas, the targets were set according to the 15 years model at appraisal. But the actual amount and the pace of achievement was much more modest than the targets. For urban areas, the targets were set at appraisal even lower than those of the 10 year and 15 year models. However, the actual achievement was still consistently below what was forecasted at appraisal in spite of the increasing emphasis on urban cadastre work for the last few years. 7.2 Supervision: 7.2.1. In general, the Bank's supervision effort was satisfactory, considering the enormous difficulties due to the security situation and communication problems. The Bank team has responded to many issues arising during project implementation and provided the borrower with practical suggestions such as postponing the cost recovery and including DGDN/CF in the institutional strengthening. The Bank's supervision mission was also effective in providing general suggestions on project management, such as formation of training teams through training of trainers program and introduction of quality control systems. 7.2.2. The Bank's supervision team faced difficulties in supervising the Project due to security problems from 1994 to 1996. The supervision mission either occurred in Paris or through the help of a local consultant. Although the project involved field work all over the country, it was difficult to follow up on the progress of cadastre work in the field. In spite of these difficulties that hindered supervision missions to Algeria during these years, the Bank's team (HQ staff and the local consultant ) was proactive in keeping constant communication not only with the Algerian stakeholders such as ANC and DGDN/CF, but also with GTZ (the co-financier of the project) to keep monitoring project implementation. The arrangement to hire local consultants to collect information needed to supervise the project worked out well. The local consultant, in a situation where the Headquarter staff could not visit the country, served as a crucial communication bridge between the Bank and the Algerian authorities. The mid-term review for the project was carried out by this local consultant in October 1996. The review focused on: (i) decentralization of administrative management; (ii) staff recruitment policy; (iii) training programs; and, (iv) achievement in rural and urban mapping. In this review, among other things, it was observed that ANC was strengthening its institution through following targets set in the SAR for recruitment of staff and a training program consistent with the recommendations from the management study in 1994. On the other hand, it was observed that institutional strengthening of DGDN/CF needed more attention. The Bank's supervision has been consistent in following-up the above concerns raised in the Mid-Term Review. Although it was initiated only close to the loan closing, a unit in charge of training programs was set up in DGDN and the joint teams of ANC and DGDN/CF in cadastre work were formulated. The Bank's supervision was instrumental in encouraging increased collaboration between ANC and DGDN/CF along with the on-going decentralization of both agencies. 7.2.3. The Bank team also effectively assisted the Borrower in monitoring the progress on the technical assistance, discussed the interim outcome of the technical assistance with GTZ, and ensured the quality of the final outputs of the various studies under the project. The supervision team was consistently proactive to this end. GTZ will continue providing assistance on the cadastre system in the coming years, particularly in the area of establishing information systems for cost accounting, quality control and land information (archiving). However, the Bank could have promoted more policy dialogue with the Algerian authorities in an effort to ensure the efficient cadastre system would be linked to the achievement of economic reform from the perspective of establishing fiscal cadastre and land valuation systems. 7.3 Overall Bankperformance: Satisfactory. - 10-  Borrower 7.4 Preparation: Borrower performance during preparation was considered satisfactory. 7.4.1. Through project preparation, the Borrower had become aware of the duplication of tasks of ANC and DGDN/CF, and the needs to simplify the procedures. During project preparation, the Government of Algeria had agreed to integrate the ANC and CF at the local level in order to improve the efficiency of title registration process, particularly in the process of updating mapping information required for title registration. Although this did not happen until close to the end of the project, it was indeed a major step forward to integrate the services of the two agencies and improve the coherence of the land titling procedure. 7.4.2. In addition, ANC and DGDN/CF, under the support of Ministry of Finance took the important steps to prepare the project before effectiveness, including the preparation of a four-year implementation plan with budget, the appointment of a project team within ANC and the recruitment of 12 public accountants by ANC. One of the efforts by DGDN to be noted is in simplifying the provisional adjudication system in which the five-year period of allowing possible claims was reduced to two years. This required five-year pending period was regarded as an obstacle to the development of a market of private land. The simplification of the process shortened the overall period of cadastre implementation. 7.5 Government implementation performance: 7.5.1. The Government's performance during project implementation is considered unsatisfactory. GoA's policy commitment and its compliance with the legal covenants were consistent during the whole duration of the project. However, during the course of implementation, the Government could have taken actions to resolve critical constraints faced by the Project such as ; (i) ensuring budget for counterpart financing as committed in the Loan Agreement when the oil price declined between 1997 and 1998; (ii) improving coordination with municipalities (APCs) in districts (wilayaete) and (iii) simplifying the disbursement procedures to facilitate project implementation and to cut unnecessary bureaucratic constraints. In addition, the implementation of Component 3 was delayed due to the poor coordination with the Ministry of Agriculture and Ministry of Justice. 7.6 Implementing Agency: 7.6.1. The implementation agencies (the ANC and the DGDN/CF) were considered satisfactory. I)uring project implementation, both ANC and DGDN/CF were committed to the project's development objectives (refer Section 5.3). Although procurement was delayed often due to administrative complications and lack of counterpart financing, the procurement was conducted in compliance with the Bank's procedure. Audit was carried out in compliance with the Bank's procedure. 7.7 Overall Borrower performance: Satisfactory. Although the Government's performance lacked in due diligence to ensure the timely implementation of the project particularly when the project suffered from lack of counterpart financing, implementation agencies made their best efforts in achieving the project's development objectives and quantitative targets. Overall Borrower Performance is thus rated as satisfactory. 8. Lessons Learned The lessons learned through this project can be summarized as the following points; Institutional Analysis. In designing a project aiming at establishing a system to deliver public sector services, such as cadastre, well-designed institutional analysis should be carried out in order to: (i) assess the current institutional set-up and human capacity; (ii) confirm the interest of beneficiaries; and, (iii) identify the optimal arrangement. Such an analysis should also include formulation of an overall implementation schedule/timetable and monitoring indicators based on realistic assumptions. In the case of this project, the focus should have been on establishing a solid base of operations rather than aiming at perfecting the system from the beginning. - 11 -  Implementation Period. It often requires ample time to set out a new system in any country. In this context, the original implementation period, 4 years, was unrealistic. The project required three extensions, which led to the duration of eight years compared to the original duration of four years. Reviews of the Bank's projects on land titling also point out that "the projects usually extend for periods of between five to seven years and form the start of longer-term government land titling and cadastre programs." Another review of land titling projects was conducted in 1992. In this review, it is reported that out of 12 operations, most performed poorly because of: (i) lack of political support; (ii) conflicting bureaucratic priorities; (iii) lack of institutional capacity of support; and (iv) complex multiple objectives with land titling as only an adjunct. Project design approach and the speed of implementation. In the project design which is focused on the establishment of a certain system of service provision, particularly projects on cadastre systems, a "building-block approach" or an incremental approach may be applied. Such an incremental approach seeks to gradually improve the system. As the review of the Bank's titling project in ECA countries states, "Implementation should not be dominated by efforts to create the perfect solutions right at the beginning of the process. The "building-block approach" may be more practical, thus enabling the project to set more realistic goals that can be met". In the case of the Cadastre Project, it was more practical to start with the two agencies, although this was not an ideal situation to start with. But by the end of the project, the two have established stronger collaboration to produce titles. Closer institutional ties and partnership are foreseen with more joint activities not only in the area of title production but also in the area of training/capacity building. Need for sensitization and outreach for beneficiaries. For the project's benefits to penetrate to the target population (intended beneficiaries), it is crucial to assess the level and nature of their interest and their potential benefit. Based on this assessment, it would be useful to consider additional efforts to sensitize them on the benefit they can obtain from title registration. One of the broad/long-term objectives of the project was to improve agricultural productivity and creation of a viable land market through ensuring the legal boundary of properties. Title registration was regarded as crucial to establish the legal base of properties. It has been proven in the project that APC/municipalities with a public campaign on absenteeism (i.e. absence of land owners' presence in the surveyed areas) have been progressing better than those without a campaign. Although negative impact of worsening security situations on absenteeism posed a constraint in field operations, importance of outreach and sensitization activities should be considered. The Project improved the number of titles made ready for the pick-up and use of the individual property owners. After the project's implementation, it is expected that the turn out of individuals requesting title registration would improve. Since the policy for registration system is developed in such a way that the delivery of titles remains dependent on the voluntary turn out of individual owners, without their turn-out, it would be difficult for the project to achieve its utmost intended benefits to increase the number of titles delivered. To create a supporting environment for the project's sustainability, it may be useful to consider additional efforts (such as publicity campaign) to sensitize people on the benefits and the procedure of cadastre. Simplification vs Complexity of existing situation in the institutional arrangement. It would be important to judge feasibility and relevancy in the case of working with multiple agencies concurrently and to reflect them in the project design. The project had to deal with the complexity of involving several agencies under different ministries and decision-making processes. It would be clearly easier to involve a single agency for the project and this was also one of the recommendations in the Final Executive Project Summary (FEPS) Review meeting. According to the review of the Bank's land titling projects and other international experiences, it is also recommended that the system of cadastre and land administration be operated by one public authority and not split between two or more ministries or authorities. In the case of the Cadastre Project, it was more "realistic" to work with the two existing agencies (ANC and DGDN/CF) and to select ANC as the main coordinator than to limit the operations to a single agency. But this has resulted in slower than planned progress in project implementation. Enhancing policy dialogue. Opportunities for policy dialogue could have been encouraged more during project implementation, particularly to increase awareness of policy-makers of the importance of cadastre/title registration linked to economic and housing policy reform. Title registration has close linkage to the establishment of a land - 12 -  tax system, and related systems such as land valuation/appraisal system. During the implementation of the project, the policy environment was still undeveloped. In addition, cost recovery through fees which were envisaged at the appraisal of the Project have not materialized yet. The initial cadastre is considered as a public service. Thus cost recovery will only materialize for transactions after the establishment of the initial cadastre. Cost accounting data will then help define fees to be charged to users. The GTZ project that follows after the completion of the Project, is considering the inclusion of an analysis on economic impacts of cadastre and title registration with a view to establishing cost recovery systems in the future. After ten years of implementation, the environment is finally beginning to favor discussion of the policy issues as above. It would be important for the Bank to continue policy dialogue with the Government on broader policy issues that ensures the sustainability of the cadastre system. Collaboration and partnership with the bilateral agency. One of the reasons for the project's satisfactory outcome was a well-coordinated collaboration/partnership with GTZ. The funding of equipment and technical assistance (training activities) were coordinated in a complementary and constructive manner, instead of overlapping of efforts. Constant communication and supervision were conducted in a mutually beneficial and complementary way (the Bank task manager attended the review meeting of GTZ, and the representative of GTZ attended the Bank supervision mission meetings). The shared objective of the two institutions was also beneficial to the ANC and DGDN/CF. GTZ, as a donor and technical assistance agency, monitored closely the implementation and also provided tools for developing a solid implementation framework for the project, such as ZOPP methodology and the logical framework. Whereas the Bank, as a financing institution, ensured the implementation process through financing not only the equipment but also training on management and technical skills. The Bank team also contributed to the continuation of dialogue with the agencies and ministries involved. The collaboration between the Bank and GTZ also occurred in the area of quality control and cost accounting systems. In this sense, the collaboration in the project maximized the strengths of both institutions in serving the borrower. Importance of human resource development and capacity development. For a project which involves institutions delivering services, it is important to address human resource development and the improvement of management skills. The training programs and the process of planning follow-up activities are an entry-point to discuss modes of closer collaboration between ANC and DGDN. The last part of the project (since 1998) shed light on the training on management skills for the top management of ANC and DGDN. This came with the understanding among all the project's stakeholders that without good communication and management skills (such as delegation of tasks), it would be difficult to meet the intended organizational goals. Although this was not envisaged at the stage of appraisal, the management of training and the formation of the training team in ANC and DGDN were found to be beneficial. They will enhance the capacity building of staff concerned and will address the organizational issues that need improvement and change. 9. Partner Comments (a) Borrower/implementing agency: PRESENTATION DU PROJET DOCUMENTATION FONCIERE GENERALE 1-CONTEXTE GENERAL ET GENESE Le projet Documentation Fonciere Generale (DFG) identifie en 1989 est l'une des deux principales mesures prises par le gouvernement pour donner un nouvel elan aux op6rations cadastrales et a I' immatriculation des proprietes foncieres sur 1'ensemble du territoire national. - 13 -  2-APPROCHE INTEGREE Le projet concu au depart pour repondre aux besoins de l'Agence Nationale du Cadastre(ANC) a et6 6tendu aux Conservations Foncieres (CF) chargees de la partie juridique (immatriculation des immeubles cadastres au livre foncier), ainsi qu'aux d6partements ministeriels et institutions de soutien direct ,a savoir l'Institut National de Cartographie et de T6l6detection (INCT), le Minist&re de la Justice(MJ), le Ministere de l'Agriculture(MA) et le Centre National des Techniques Spatiales (CNTS).Les moyens identifies couvrent 1'ensemble des besoins tant sur le plan technique que juridique, institutionnel et organisationnel. ll a te juge utile dans ce cadre d'operer, au pr6alable, une simplification du dispositif juridique et institutionnel regissant le cadastre general et le livre foncier ainsi qu'une r6organisation appropriee des services concern6s.Ces changements se sont traduit par: -l'amendement du decret n