Algeria - Cadastre Project

Z-A. 3a/+ - d-e Document of The World Bank FOR OFFMCIAL USE ONLY MICROFICHE COPY Repwt No. 10346-AL TEepcrt No. 10346-AL Type: (SAR) LACOUR, V / X37335 / H4169/ EMEN1IN STAFF APPRAISAL REPORT DEMOCRATIC AND POPULAR REPUBLIC OF ALGERIA CADASTRE PROJECT MAY 21, 1992 Infrastructure Operations Division Country Department I Middle East and North Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents way not otherwise be disclosed without World Bank authorization.  CURRENCY BOUIVALENTS (As of March 1992) Currency Unit = Algerian Dinar (DA) US$1.00 - DA 22.50 DA 1.00 USS 0.044 ABBREVIATIONS AND ACRONYMS ANC Agence Nationale du Cadastre (National Cadastre Agency) APC Assembl6e Populaire Communale (Municipality) BA Banque d'Alg6rie (Central Bank) OCEP General Cadastre Establishment Program 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 G6n6rale du Domaine National (General Directorate of Lands) EAC Exploitation Agricole Collective (Privatized Collective Farm) EAI Exploitation Agricole Individuelle (Privatized Individual Farm) EPA Etablissement Public Administratif (Autonomous Public Agency) EPIC Etablissement Public a Caractere Industriel et Commercial (Commercially managed Public Agency) GDP Gross Domestic Product GFCF Gross Fixed Capital Formation INC Institut National de Cartographie (National Mapping Institute) INF Institut National des Finances (National Finance Institute) ITTPB Institut Technologique des Travaux Publics et du Batiment (Institute of Technology for Public Works and Building Construct ion) LCM Law on Money and Credit LIS Land Information System MOH Ministere de l'Habitat (Ministry of Housing) MOA Ministere de l'Agriculture (Ministry of Agriculture) MOE Ministere de l'Economie (Ministry of the Economy) MOJ Minist&re de la Justice (Ministry of Justice) FISCAL YEAR January 1 to December 31  FOR OmCIAL USE ONLY DEWMORTIC ADPOPJUL REPUBLIC RE ALGEIA CARDASTRZ PROJUt= STAFF APPRAISAL REPORT Table of Contents Loan Su mary . . . . ......................................... . . . I MNTRODUCTION . . . . . . . . . . . . . 1 II* LAND RESOURCE US AND REGISIRATION 1. . . . . . . . A. Land and the Economic Reforms 1 B. Land Use and Policies 3 C. The 1975 General Cadastre Law and Its Implementation 7 D. The Now Institutional and Procedural Setting 10 E. The Remaining Agenda 15 III. THE PROJECT . . ... o .. 15 A. Government Strategy 15 B. Rationalo for Bank Involvement 15 C. Project Objectives and Summary Description 16 D. Cadastral Mapping and Parcel Adjudication 16 S. Strengthening of the National Cadastre Agency 19 P. Technical Assistance, Training and studies 20 IV. ORGANIZATION AND IMPLEMENTATION ..... . . .. . . . . .. . . . . . 22 A. The National Cadastre Agency 22 B. Project Implementation Arrangements 23 C. Monitoring, Evaluation, and Reporting 25 V. PROJECT COSTS AND FINANCING ..26 A. Project Costs 26 B. Project Financing 28 C. Procurement 28 D. Disbursement 29 E. Cost Recovery 30 F. Accounts and Audits Requirements 30 VI. PROJECT JUSTIFICATION AND RISKS . . . . . . .. . ..31 A. Benefits 31 B. Risks 32 C. Environmental Impact 32 VII. AGREEMENS AND RECONNENDATION ... . . . . . . . . . . . . . 32 This report is based on the firdings of an appraisal mission which visited Algeria in Novesmer 1991. The mission comprised of Messrs. Vincent Lacour tsenior Housing Specfatist), Mostafa EI-Erian (Lawyer); Nabil Shehadeh (Principal Financial Analyst); Francis DOelbecque iConsultunt, Training Specialist); Jacques Gastoldf (Consultant, Cadastre Specialist) and Bernard Samoullier (Consuttant, Mapping Specialist). Charles Sterling (Operatfons Analyst) and Pierre-Etfewme Bouchsud (Consultant, Economist) assisted fn the preparation of the report. Task Manager: Vincent Lecour (MMIIN); D.vfslon Chief: Amir AL-Khafaji (CNIIN); Director: Pfater P. Bottelier (MI1); Peer Revfewer: Maurice Ass o CAFlAG). 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.  Table of ContentE (cont'd) ANNEXES . . . . * . . . . . . . . .*. .*. .*. 34 1. Cost Estimates and Financing 34 2. Technical Assistance, Training and Studies 42 3. Cadastre Production costs 57 4. Performance Indicators and Annual Targets 59 5. Supervision Schedule 60 6. Project Implementation Timetable 61 7. Disbursement Schedule 62 8. Documents Available in the Project File 63 FIGURES 1. Algerian Population and Space 3 2. Cadastre Establishment Procedures 8 3. Cadastre Maintenance Procedures 10 4. Institutions 12 5. ANC Organization Chart 22 6. DGDN Organization Chart 24 TABLES 1. Pilot Zones 18 2. Project Cost Summary 26 3. Project Cost Summary by Category 27 4. Project Inflation (percent per yearl 27 5. Financing Plan (US$ Million) 28 6. Project Expenditure by Procurement Category 29 Hap No. IBRD 23573  RENOAIC AND POPULAR REPUBLIC OF ALGERIA CADASTRE PROTECT STAFF APPRAISAL REPORT Loan Summarv Borrower: Democratic and Popular Republic of Algeria Beneficijriest Ministry of Economy: General Directorate of Lands, National Cadastre Agency (ANC) Amount: USS33 million equivalent Terms: Repayable in 17 years, including a five-year grace period, at the Bank's standard variable interest rate Descripti2a: By supporting 'plementation of the cadastre, the project would help thc 0overnment of Algeria to create a legal and institutional framework to ensure a functional land market; to foster private investment in agriculture, housing, and industry; and to lay the foundation for a sound land-taxation policy. Project objectives are to: (a) strengthen the institutions that implement the cadastre; (b) support an increase of annual land registration over four years; and (c) develop cadaotre-based information systems for taxation and other land-related uses. The project would include three components: (a) a four-year cadastral mapping program for 3 million ha of rural land and 36,000 ha of urban land, during which mapping techniques would be tested, and computerized land registration and map production would be introduced; (b) strengthening of the ANC; and (c) provision of project-related technical assistance, training, and studies. Benefits: The main benefits of the proposed project are a strong contribution to private sector development through greater individual security of land tenure and a new, accurate cadastral map of Algeria's densely populated usable land. The current lacic of these as well as a properly functioning land market make project benefits impossible to quantify at this time, but implementation of the project is expected to provide the basis for quantitative valuation in the future.  - Li - Preliminary information suggests that, on average, titled rural land is worth 60 percent more than undocumented land. Risks: The project's main risk is that the institutional strengthening component may not be carried out at a sufficiently rapid pace to enable ANC to use efficiently its Lncreasing technical capacity to implement the GCEP. This risk is minimized by making the cadastre as simple as possible, setting it within a strong institutional framework and defining performance indicators and targets which will be monitored through performance plans to be agreed upon by the Government and ANC.  - iii - nR3IOC3AY!C AND POPULA REPUDIC OF ALG3RIA OMas PROJECT 3stisated Costs and Financing Plan Us$ Million ______________________ Local Foreign Total ______________________ ESTIMATED COSTS A. Cadastre map & parcel adjudication 1. Aerial photography 2.6 7.1 9.7 2. Digital stereoplotting 1.9 3.2 5.1 3. Ground survey 23.3 8.9 32.2 4, Parcel adjudication 10.6 1.S 12.1 5. Digital mapping 1.4 2.6 4.0 6. Legal data base 0.8 1.5 2.3 Subtotal 40.6 24.8 65.4 B. Institutional support to ANC 1. Training 1.6 0.6 2.2 2. Organization strengthening 1.0 2.7 3.7 Subtotal 2.6 3.3 5.9 C. Technical assistance & studies 1. Studies for land administration 0.2 0.4 0.6 2. Land inwormation systems 0.3 0.5 0.8 Subtotal 0.5 0.9 1.4 TOTAL BASELINE COSTS 43.7 29.0 72.7 Physical Contingencies 4.4 2.9 7.3 Price Contingencies 13.4 2.6 16.0 TOTAL PROJECT COSTS1/ 61.5 34.5 96.0 FINANCING PLAN Government 61.51/ 1.5 63.0 IBRD 0.0 33.0 33.0 TOTAL 61.5 34.5 96.0 1/ Includes US$23.6 million of taxes and duties.  I. INTRODVCTI 1.1 Although the most important breakthrough in private sector development that Algeria has accomplished in its ongoing economic reform concerns land ownership, its effectiveness is still hampered by the lack of a land right- ' .ta base. Land ownership and transactions used to be highly restricted ;nsistent with the central planning that was used to manage the economy. I parallel with the economic reforms, fundamental changes in land law are takLag place. Before the beqinning of the economic reform process, which has been progressing since 1988, experiments in agricultural land privatization were carried out as early as 1983. Between 1987 and 1989, most public agricultural land had been assigned to privatized groups of farmers and the municipal monopoly on urban land transactions removed. However the Government of Algeria has yet to complement these sweeping changes by giving titles to the new owners, thereby enabling them to carry out land transactions, and by establishing a proper land-rights data base. 1.2 The proposed project is designed to set up a land-rights documentation system (cadastre) that would enable smooth functioning of the emerging land market and foster land-related private investment. This cadastre would include a cadastral survey of urban and rural property boundaries and a register of titles that together would make legal information on land rights easily available and simplify conveyancing. Project objectives are to strengthen the institutions in charge of cadastre implementation and maintenance, to support the progressive increase in cadastre annual production and maintenance within a four-year time slice -- 3 million ha of rural land and 36,000 ha of urban land would be mapped and 1 million parcels adjudicated -- as a contribution to the Government's program to complete the cadastre in the northern part of the country within 15 years and to develop cadastre-based information systems for taxation and other land-related uses. 1.3 The proposed project was identified in late 1989, and prepared by a study carried out between October 1990 and June 1991 and by a series of missions culminating in substantial streamlining of the legal and institutional setting of the -adastre. It was appraised in November 1991. II. LAND RESOURCE USE AND REGISTRATION A. Land and the Economic Reforms Economic Reforms 2.1 Land privatization, although still incomplete, has been the first, as well as most dramatic, move Algeria has made toward creating a market economy. However, before the private sector can benefit fully from the ongoing reforms, a well-established land-registration (cadastre) system must complete the land management framework. The reform program that the Government is undertaking began in 1987 when it decided to dismantle the'state farm sector, which represented 40 percent of agricultural output and to give usufruct rights to about 27,000 private groups of farmers. In 1989, the Government launched a second reform package to replace public enterprises'  - 2 - centralized planning by a decentralized management structure. In 1990 the Law on Money and Credit I (LMC) was issued; this law established the central bank (Banoue d'Alafrie, BA) as an independent institution with responsibility for setting monetary policy and overseeing the banking system. The LMC contributes to leveling the playing field between public and private enterprises and provides the legal framework for a market-driven financial sector. In the same year, the Land Policy Law was adopted, which provides for privatization of urban land transactions. Although many components of the legal framework for a land market are now in place, a full-fledged market d aling with various types of rights such as ownership, lease, tenancy, and mortgage cannot function yet becauLe of the lack of widely recognized evidence of these rights. Country Overview 2.2 Management of Algerian land has reached a critical stage for the Algerian economy because of three main factors: (a) a rapidly growing population; (b) heavy dependence on food imports; and (c) the modest share of usable land in the national territory. Algeria has a population of 26 million, which has been increasing rapidly at 2.7 percent per year; the urban population, which represents 40 percent of the population, is growing faster at 5 percent per year. Housing and feeding this growing urban population are among the most serious ^hallenges to the country's development. 2.3 Heavy reliance on food imports, compounded by the very narrow range of export commodities, makes Algerian food policy vulnerable. Imports represent a large part of national food consumption3' and are financed by export income, of which crude oil and gas represent 96 percent. Gains in agricultural production would have a positive impact on the current account. 2.4 Usable land is scarce in Algeria, for rural as well as for urban needs, and competition between these two uses is intense. Although Algeria is the second largest African country with 2.4 million km2, because of limited rainfall, only 75,000 km2, or 3 percent of the territory, is suitable for cultivation; 400,000 km2 are considered suitable for grazing, foxests, esparto scrub, and shrubs. The Algerian population is also concentrated in a small portion of the national territory; 80 percent live on less than 8 percent of the territory (Figure 1) along the northern coastal strip. This area contains almost all of the country's assets: deep soils, abundant rainfall and irrigation potentialities, as well as communication infra- structure and large markets. Although the Government has constantly tried to foster urban settlements in the less fertile region of the high plains, this policy has had little influence on the urbanization trend. Therefore, an important item of the land management agenda is to promote an efficient use of urban land. 2/ Loi 90-10 relative A la monnaie et au cr6dit, April 14, 1990. 3/ Tie self-sufficiency rate is 25 percent for cereals and 40 percent for milk.  -3 Figuro It Algerian Population and Space Algerian Population by decreasing density 10o%- -.-..--.-. 90% - - -- 80 -- 70% - 60% 2 50%. 0. 30%-- 20%-- 10%-- 0% - _ _ 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% area S. Land Uns_ and Policies Allocation of the scarce arable land is being rationalized 2.5 Most of Algeria's rural land is located most in the northern part of the country. The 7.5 million ha of arable land can be broken down from north to south into strips parallel to the coast; they consist of: (a) 550,000 ha (of which 220,000 ha is irrigated) in a hot climate with abundant rainfall, on which citrus and early vegetables are intensively cultivated; (b) 5 million ha in plateau areas, where winters are cold, rainfall fluctuates sharply, and cropping consists mainly of a wheat or barley and fallow rotation; and (c) 2 million ha situated between the dry south and the desert, and in the mountains, where cereal cropping is irregular. 2.6 Until 1987, when the state farms (Domaines Aoricoles Socialistes, DAS) were dissolved, these relatively large-scale, capital-intensive farms covered a large part of the more fertile land in the higher-rainfall areas.  -4- When Alga&ia became independent in 1962, agricultural land belonging to French settlers, which was generally located in the rich coastal strip, was collectivized into large state farms averaging 1,000 ha in size. In 1971, in a second step known as the Agrarian Revolution, the size of private property was limited to about 80 ha per family and excess land was incorporated into state farms. In 1987, the DAS numbered about 3,100, had an average size of 800 ha, and covered a total area of 2.5 million ha, or about one-third of total arable land. Because the farms were in the more fertile area, however, the propo.,rtion of their land left fallow was much small.er than in the private sector, so that land cultivated by the DAS represented 40 percent of total area cropped in any year. This sector contained about 60 percent of Algerian production potential. It employed about 140,000 people as full-time DAS workers anxd hired part-time workers totalin.; about 180,000 (full-time equivalent). 2.7 Private farms, which cover two-thirds of the arable land and account for 85 percent of the agriculture wo_k force, represent only 40 percent of the Algerian production potential. About 60 percent of the 850,000 private farms are on less than 3 ha, although it is likely that many farmers operate more than one farm. This sector cultivates only about 60 percent of the arable area it occupies because much of the land falls in marginal areas and is left fallow. By contrast, in the livestock subsector, which plays an increasingly important role in agricultural production, the private sector dominates; about 95 percent of the animals are privately owned. 2.8 A major change occurred in 1987, when the Government passed a law4/ providing for the division of DAS into autonomous groups of private farmers. The breakup of the DAS involved the distribution of land and assets to voluntary associations of no fewer than three farmers. About 22,000 cooperative farms (Exoloitation Acricole Collective, EAC) were established, each occupying an average of 80 ha of cultivable land with 6-7 members. Small parcels of land left over were allocated to individual farmers (Exploitation Aaricole Individuelle, EAI)l 5,100 EAIs were also established, each with an average size of about 6 ha. 2.9 The 87-19 law introduced private management in former state farms, but it strongly limits the right of EAC and EAI members to dispose of their land and prevents a formal land market from emerging. Because Government officials believed that small-scale farming operations would be inefficient, this law states that state farm land should generally be broken up into EACs of at least 70-90 ha. EAC and EAI usufruct rights cannot be sold during the first ten yf:.'re following their constitution, and their members cannot rent their right. (these rights may be forfeited in case of land rental or use of land for non-agricultural purposes). These are obstacles to optimizing land use and to consolidating parcels. There is ample evidence that an illegal land rental market is developing. 2.10 Although the initial breakup of the DAS was carried out in a 4/ Loi 87-19, December 8, 1987.  remarkably short time, some further adjustments were necessary and are still taking place. This reform has been carried out in three steps. First, the DAS were broken up and reassigned to cooperative and individual farms between October 1987 and May 1988 on a first-come, first-served basis. Second, between 1988 and 1991, 380,000 ha were given back to their owners from before the 1971 expropriation; the unfair control of land allocation and claims expressed by the farmers whose land had been nationalized in 1971-73 led to major amendments to land allocatlon. Third, dissatisfaction among EAC and EAI members led to an ongolng reallocation of plots. As a result, the number of EAIs has more than doubled (currently 15,900); the average EAC membership has decreaoed to 5.5, and thelr number has increased to 25,400. 2.11 At thLs last stage of the implementation of the 87-19 law, the legal framework proved too rlgld and it was determined that: (a) most EACs were unfeasible and should be broken up into individual farms; and (b) EAC and EAI farmers should be given more leeway to reshape their farms. Cooperative farms were often created without beneficiary willingness to share in operations management. A growing number of farms ara being split into individual farms, although the legal basis for these individual farms is precarious. Experience gained ln other countries shows that individual farms are more profitable than cooperative farms. Agricultural operations could reach the optimal slze by buying or renting supplementary land, which the legal framework does not yet allow SAC and EAI to do. Municipal control of the use of urban land is relaxed to give more space to market forces 2.12 A large share of urban areas lies either in the most productive agricultural belts or in piedmont areas, unsuitable for construction for geotectonic reasons. Some 80 percent of the population lives in the coastal area, and three of the four major cities (Algiers, Oran, and Annaba) are maritime ports. Urban land represents about 300,000 ha, and it is estimated that urban population growth (5 percent per year) generates the need for an additional 15,000 ha per year. 2.13 The Urban Land Reservation Law';, which the Government issued in 1974, established an urban perimeter for every city, prohibited building construction outside thls perimeter, and entrusted the municipalities (Assembl6es Populaires Communales, APCs) with providing suitable land for urbanization. Provision of land suitable for urban growth and the preservation of agricultural land were the objectives of this law. Within the Urbanization Perimeter, state land was transferred to the APCs and private undeveloped land could be sold only to APCo, which, in turn, were supposed to provide land for public infrastructure and housing construction, as well as for private individual housing construction. 2.14 Implementation of this law was unsuccessful and has resulted in inefficient use of urban land and encroachment on agricultural land. APCs, 5/ Ordonnance sur les r6serves fonciares No. 74-26, February 20, 1974.  which had a monopoly on land supply, had insufficient expertise, management capacity, and fundu to provide adequate land to meet needs stemming from urban growth. APC used to allocate plots to construction projects during ad hoc meetings, but conveyancing6/ was rarely carried out. Prices were fixed by nationwide rules that did not represent the economic value of land and, because the pricce were seldom paid, did not constitute an incentive for efficient use of urban land. Because land-use control was more stringent in the urban perimeter, the growing demand for urban land resulted in increased encroachment on agricultural land outside this perimeter. By the end of 1989, out of 564,000 plots that had been allocated to households for family house construction, 170,000 were owned by households that did not plan to build in the near future. Simultaneously, 300,000 illegal housing constructions had been tallied in the agricultural land surrounding the main cities. 2.15 The Land iolicy Law7/, which the Government passed in 1990, dissolves the Urban Land Reservation System and provides for free land transactions. APCs have thus lost their monopoly on urban transactions and prices are freely determined by the market. Land is still segmented into urban and agricultural zones, however, and construction remains firmly restricted in the agricultural zones. Zoning maps are to be prepared under APC supervision, and APCs can monitor the evolution of the urban land market by exercising their authority to preempt land. Land-management agencies should be created to this effect. The lack of legal titles is hampering the implementation of these new policies 2.16 The lack of legal titles in rural as well as urban areas prevents the land market and private land-related investment from developing. Only 5 percent of private rural and urban owners have legal evidence of their property rights. In the former DASs, only 5 percent of the government titles (actes administratifs)8/, have been issued. However, there are already some indications of improvement: 20 percent of government titles for EACs and EAIs are currently beinj processed and, because composition of EAC/EAIs is stabilizing, it is expected that title issuance will now be carried out at an accelerated pace. In urban areas, most of the transactions carried out since 1974 were never translated into proper conveyancing. As a provisional registration system, the Land Policy Law stipulates that APCs will have to make a land inventory that will be progressively replaced by the legal registration system to be provideG by the cadastre. 6/ Conveyancing is the procedure by which the transfer of property from one person to another is performed. 7/ Loi d'Orientation Fonci6re No. 90-25, November 18, 1990. 8/ Government title, (acte administratif), is the particular title defined by the 87-19 law as legal evidence of EAC/EAI beneficiaries, rights.  - 7 - Land taxation is not conducive to the development of the land market 2.17 Although land ownership and use is virtually untaxed, taxes related to land transactions are heavy and are an obstacle to proper conveyancing and to the development of the land market. Taxation of agricultural land was abolished in 1972, and schedules for urban land taxation have not been updated since 1934; new constructions are exempt for 10 years; moreover, urban land tax collection is poorly administered and generated an annual income of about DA 300 million in 1988 (or .3 percent of budgetary income). In contrast, land transactions are heavily taxed; transaction tax rate is 16-20 percent of land value plus about 2 percent for recordation. The 1992 Annual Finance Law, which indicates changes in land taxation policy, has yet to be complemented by practical measures. C. The 1975 General Cadastre Law and Its Implementation 2.18 Land registration is not new to Algeria. It has existed since 1863 but was limited to a small part of the national territory and has not been maintained for the last 15 years. Before French colonization, individual ownership was uncommon because land transactions were limited by religious (habou or wakf land) or civil (arch or melk land) rules. Land registration was introduced in 1863 by the French Government to enable settlers to obtain land and protect their property. Based on the Torrens system, it did not cover the whole territory, but concentrated on wealthy rural and urban estates. In 1962, many French settlers vacated their properties, which were either confiscated by the State or acquired by individuals. Although land registration was not abolished, it became progressively outdated. Since 1971, it has been further disrupted by nationalization resulting from the implementation of the Agrarian Revolution and by the Urban Land Reservation Law. Little attention had been given to conveyancing during the nationalization process. 2.19 A comprehensive legal framework for land registration9/ was defined in 1975 and 1976; it has been scarcely implemented so far but constitutes a basis for a practical land-registration system (Figure 2). The Government's objective was to establish a General Cadastre territory north of the 34th parallel10/ in ten years. The General Cadastre, as defined by the 1975 General Cadastre Law, is based on the principles of the Torrens system. It includes a cadastral map to be established by the cadastre services and a land register to be established by the Land Registry Office (Conservation 9/ Ordonnance sur le Cadastre General et le Livre Foncier No. 75-74, November 12, 1975 and D6crets 76-62 et 76-63 relatifs a l'6tablissement du Cadastre G6n6ral et a l'institution du Livre Foncier, March 25, 1976. 10/ This portion of Algeria is the active part of the territory for land management purposes; it represents about 110,000 km?.  Figure 2: Cadastre Fstablishaent Procedures National Mapping Institut | National Cadastre Agency || adegisirZJ | ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~I lWra Cdoto L| m w~~~~d r'ed s. 1"I 0'..V .-..Y ~ ~ ~ ~ ~ ~ ~ p...a,.di,* L:.I dtX Base flap .Ifts- Fonciere, CF). The cadastral map describes the parcel boundaries. The parcele of a section11l are numbered and listed on a cadastre register where the names of the apparent owners12/ are indicated on the basis of the results of the investigation carried out by a cadastral commission. This commission is chaired by a local judge and includes local government representatives and cadastre officers; it can settle minor disputes between neighboring landowners. Once the cadastre investigation and survey are completed in the whole commune, these documents are passed to CF, which carries out its own investigation and delivers titles, which thereafter 11/ A section consists of the parcels that can be represented on the same map; it ranges from 5 ha in dense urban areas (scale 1:500) to 500 ha in rural areas (scale 1:5,000). 12/ An apparent owner is a person who appears to have a strong interest in the parcel according to the initial evidence gathered during cadastre investigation; the cadastre register does not constitute any evidence of land rights.  - 9 - constitute the legal evidence of land rights. These titles are recorded in the land register. Accuracy of the land titles is guaranteed by the State. Maintenance of the cadastral map, land titles, and land register are entrusted to the cadastre services and to CF. Lessons drawn from cadastre implomentation 2.20 The very modest implementation of the General Cadastre after 16 years' experience reveals that its design was too ambitious to be practical and that the institutional setting was inadequate. In the territory targeted, less than 5 percent of rural land and less than 1 percent of urban land have been cadastred. In the communes cadastred, the cadastral map and the land register have been improperly maintained. Project preparation has demonstrated that cadastre implementation was constrained by lack of institutional accountability, inadequate cadastre establishment procedures, and unreliable conveyancing and cadastre maintenance. The institutional setting was not adapted to the completion of cadastre establishment in a reasonable period of time and did not ensure that this task would be completed efficiently or with sufficient quality. The cadastre services have devoted only 20 percent of their activity to cadastre establishment and have accomplished various tasks, such as EAC subdivision surveys and public equipment surveys, that could be carried out by private surveyors. The roles of cadastre services and CF were partially duplicated (para. 2.21(b)), which created inefficiency and diluted accountability. Experience gained in other countries shows that unified leadership is critical for cadastre establishment. 2.21 Cadastre establishment procedures were too complex, as well as redundant, and did not take into account the existing diversity of land-right justifications: (a) survey methods used in urban areas were too lengthy and expensive, included extensive field measures, and aimed at an unnecessary accuracy; (b) land rights were investigated twice, first by the cadastre survey team and a second time by the CF titling team; and (c) although land titles should ultimately replace any kind of land-right justification, titling procedures have not been able to cover a significant portion of properties because they have not recognized traditional land-right justifications in a timely manner. Because conflicts about boundaries and ownership have been relatively limited (less than 5 percent of cadastred communes), it would make sense to streamline titling procedures for those parcels that are covered only by traditional land-right justifications when the ownership is not questioned.  - 10 - Figure 3: Cadastre Maintenave. Procedures Seller/ nuotry rveyor ANC |CF l ~__ _ SAIE __ _lzzm Saw$ M___ __ ___ ___ __titl Ige eih ont _ deposittIn in ti Irrswry esrow accounSi been complex and have not guaranteed sufficient accuracy, and the docWments (seven to nine) for each transaction have made the pocdueos coaepcrd MaP estalis D. The New Institutional and Procedural Setti~~nd rl-o 2.22 Conveyancing and cadastre maintenance procedures aFigure 3) have been complex and have not guaranteed sufficient accuracy, and the professionals that implemented these procedures have not been accountable. For each transabtion, CF and the cadastre service have had to intervene at different steps; the number of steps (five to nine) and the number of document (seven to nine) for each transaction have made the procedure costly and the risk of crror very important. Conveyancing professionals, such as notaries and surveyors, have been few and inadequately skilled to be able to implement conveyancing according to the land-registration system. The quality of their performance should be improved by more effective accountability and an increased participation in cadastre establishment. D. The Now Institutional and Procedural Setting 2.23 Experienceo gained in cadastre implementation and an intense dialogue between Algerian officials and the Bank during project preparation have resulted in substantial modifications, intended to brLng about better implementation organization and simpler and more efficient procedures. Remaining issues concern the testing and fine tuning of these modifications  - 11 - and laying the groundwork for their full implementation in a new General Cadastre Establishment Program (GCEP) that aims to complete cadastral mapping and parcel registration in the territory north of the 34th parallel within a period of 15 years. Institutions 2.24 The current institutional setting (Figure 4) under the auspices of the General Directorate of Lands (Direction G4n6rale du Domaine National, DGDN), within the Ministry of Economy (Minist6re de 1'Economie, MOE), gives clear responsibilities to ANC for cadastre establishment and to CF for parcel adjudicationl3l and land register maintenance. The National Mapping Institute (Institut Natlonal de Cartoaravhie, INC) is the only Algerian aerial survey producer. The institutional setting also includes the district courts that settle disputes arising from the adjudication process and can be involved in some adjudication, and the notaries and surveyors that implement the conveyancing procedures. 2.25 DGDN supervises public land management and land registration. It includes a Directora.te for Cadastre and Land Title Registration and a Directorate of Public Lands (Figure 6). It is DGDN's responsibility to prepare and mainta.n the legal framework for land registration and to supervise cadastre establishment and maintenance. 2.26 ANC was recently made responsible for implementing the GCEP. ANC was created14' in 1989 out of the Directorate for Cadastre and Land Title Registration to improve the efficiency of cadastre services. ANC is an autonomous executing agency (Etablissement Public Administratif, EPA) under the auspices of DGDN and is financed from the State Budget. It has 48 local branches at the wilavia5' level, which are locally responsible for cadastre establishment, and five regional branches. Until recently, ANC's main mlssion has been cadastral mapping using orthophotographic maps produced by INC and its own field measures. In 1992, ANC was given full responsibility16/ for the implementation of the GCEP, including the collection of data for parcel adjudication. In the coming years, ANC will have to improve its legal 13/ Adjudication is the legal action by which the rights of a land holder are recorded in the land register. once a parcel is adjudicated, the land register, or aal extract (the title), constitutes evidence of legal rights. Accuracy of the land register records is guaranteed by the State. 34/ D6cret ex6cutif 89-234 portent creation de l'Agence Nationale du Cadastre, December 19, 1989. 15/ Wilaya is the principal administrative subdivision of the Algerian territory. There are 46 wilayas in Algeria. 16/ Decret executif 92-63, February 12, 1992 and Decret ex6cutif 92-134, April 7, 1992.  - 12 - expertise and remodel its management and organization to increase its efficiency, accountability, and capacity to adapt to a rapidly changing sicuation. Figure 4: Institutions Institutions Ministry Ministry 1 InWrI hierarchical link of Justice of the Economy of Defense administrat ive supervision - - - l I |Notaries| Surv yorsl | District Land National National (600) C(200) ourts (200) Registry Cadastre Mappinq (CF) ~Agency Institute (ANC) (INC) Cadastre Establishment Settlement of Parcel Ground survey Aerial land-related adjudication cadastral maps survey disputes Preparation of parcel adjudication Maintenance of cadastral maps and land registers Conveyancing Ground survey Settlement of Maintenarnce of for parcel land-related cadastral maps boundary disputes modification 2.27 CF, which is in charge of parcel adjudication (para. 2.32), has good legal expertise but is ill equipped to face the growing workload that would result from the implementation of the GCEP. CF is part of DGDN (Figure 6) and has local branches at wilava and district (daira) levels. Currently, CF devotes most of its capacity to the maintenance of the deeds and titles resulting from the former land-registration system. Within the GCEP, CF will adjudicate parcels Based on data collected by ANC. To deal with the new titles that the GCEP will produce to replace the former deeds and titles, CF will have to increase its capacity and to improve its coordination with ANC.  - 13 - 2.28 INC iu the only Algerian aerial survey producer and is entrusted by the Government with the control of all aerial survey activities over Algerian territory. However, it can contract out aerial photographing, provided that an INC navigator is aboard and the photographs are processed at INC's facilities. INC is a commercially managed public agency (EPIC) under the Ministry of Defense. 2.29 Judges are part of the cadastral commissions and settle minor disputes during cadastre investigation (para. 2.31), although parcel adjudication is CF's responsibility. Courts are inadequately equipped to address the growing number of suits involving land rights that will be filed as a result of cadastre generalization. Creation of land sections in local courts and training of judges in land matters is envisaged. 2.30 Notaries carry out conveyancing (para. 2.35), with the participation of surveyors when boundaries are to be modified. The Government privatized the notaries in 1990 but has yet to increase the number and improve the skills of surveyors, and to a lesser extent, of the notaries. Notaries used to be civil servants whose duties and responsibilities were meager because of the low level of activity of the formal land market. Since 1990, 600 notaries have been admitted to open a notary office, and a Notary Association has been created. Surveyors, who number 200, also have a professional organization. The quality of their activity, however, is not yet controlled. It is envisaged that a quality-control system would be established to ensure the quality of title and cadastral map maintenance and enable these private professionals to play a more important role in cadastre implementation. Legal investigation 2.31 Field investigation by ANC teams is generally well done, and the participation of a judge makes it easy to solve minor legal issues arising during investigation. The second investigation that CF used to carry out for title issuance purposes added little value to the first. It was both confusing and time-consuming for the citizens concerned. DGDN has decided to strengthen the legal skills of ANC teams by the secondment of CF staff and to gather all information needed for the cadastral map as well as for adjudication in a single step. Adjudication 2.32 Adjudication is straightforward for titled parcels but takes five years for the untitled parcels, which represent more than 75 percent of the total stock. New titles are issued directly against old titles when these are still valid. There are relatively few valid titles owing to the paucity of old titles and subsequent land-management reforms. In the remaining cases the provisional adjudication system freezes all transactions for five years to allow possible claims to be filed, which is detrimental to the development of a private land market and is likely to increase the number of illegal transactions. Moreover, experience shows that objections are filed only during the last year of this period.  - 14 - 2.33 DGDN has agreed to simplify the provLsLonal adjudLcatLon system. The flve-year perLod wlll be reduc3d to two years. Cadastral teams will implement the exlstlng procedure by whlch anybody who can demonstrate a 15- year peaceful occupation of an untitled parcel becomes owner of that parcel. These eimpliflcatlons should shorten the overall period for cadastre implementation. Draft decrees concernlng these simplificatlons were revlewed during appraisal. It is estimated that ln rural araas 90 percent of titles would be ellgible for absolute or provlslonal tltle lssuance. The remaining propertles will be adjudLcated after a judLcLal declslon. Maintenance and convoyancing 2.34 Malntenance procedures are reasonably well-implemented when changes do not alter boundarles. However, thelr implementatlon during the cadastre implementatLon phase is unclaar. Because cadastre implementation can last up to four years in a single commune, this jeopardizes the reliability of the new cadastre. DGDN has agreed to clarify the maintenance responsibilities to improve maintenance in the case of boundary modifications. Moreover, the 92-134 executive decree allows ANC to promulgate the cadastre more progressively so that each cadastre section can be completed in no more than a year. 2.35 Although the new LnstLtutLonal settlng streamlines conveyancing procedures, these procedures have, indeed, to be made consistent with the development of the prlvate sector and approprlate financial instruments. The most striking feature is that before any transaction can be conveyed, its total amount has to be deposited with the Treasury and cannot be released before the conveyancing is completed. As a result of this feature and of high transaction tax rates, most transactions are carrled out without formal conveyancing. Systematic cadastre establishment and voluntary title issuance 2.36 In addltion to the systematic cadastre establishment program, alternatives allow land-rlght holders to get legal evidence of their rights. The systematic approach that was chosen by the Government is cost-effective and has proven to be successful ln many countrles. During implementation, however, it creates lnequities among holders of land rights. To mitigate these inequities, the OCEP gives priority to areas where land markets will be particularly active (suburban areas and highly productive rural areas). Alternative approaches have also been designed to enable land holders to get legal evidence when they need it to sell thelr land or lnvest ln it: (a) the notorlety deed, which was instituted by decree 83-352, is a full ownership title and can be filed voluntarily with CF; (b) all the legal deeds made before 1971 are legal evidence of ownership; and  - 15 - (c) the Land Policy Law (90-25) institute. a possession certificate (certifLcat de cossessLon) that is not a fully transferable title but can be used to obtaln a buildLng permlt or mortgage. B. The Remaiing Aaenda 2.37 The Government has recently improved the lnstitutlonal setting and procedures for the new cadastre and now ham a workable GCEP setup. GCEP lmplementation will, however, reqiLre that: (a) the procedures be fine tuned; (b) the framework for an effective land market be organized, partlcularly with respect to the notarLes and surveyors; (c) lLnks between the Government and ANC be established on a basis that would ensure long-term flnanclng and efflciency LncentLves; (d) ANC organLzatLon and management be lmproved; (e) ANC staff be tralned and surveyors be suffLcLently skllled for subcontracting to be selected; and (f) the mapplng program be contlnued. III. THE PROJECT A. Government Strateav 3.1 Land registratLon is a structural element ln the vast reform program that the Government is undertaklng. The Government ls strongly committed to having the new cadastre establlshed wlthin the next 15 years. Land registration is a key element in the agrlcultural reforms, where it conditlons the emergence of a land-market and a better allocation of land, and in the urban land reforms, where lt is vlewed as a formal land market development and monltorlng tool. It is a cornerstone of the flnancial sector reform and a pre-condition of private investment. The Government has decided to finance cadastre implementation from the State Budget and to recover only maintenance costs from the users. Although land tax reform is currently under way, it is not expected to use cadastre lnformatlon at the very beginning of its implementatlon. Rather, land tax reform would be based on declarations from land users and would use cadastre information only in a second stage when the cadastre would have reached a crltlcal slze. B. Rationale for Bank Involvement 3.2 The project is a key part of the current econom.c- reform process that the Bank supports, partlcularly for prlvate sector development. The forthcomlng HousLng and Constructlon Sector Adjustment Loan (FY93), Economic Reform Support Loan II (FY94), Local Government Development Project (FY95), and Agrlculture Sector Adjustment Loan (FY95) are closely connected to land reglstratlon. This project would also pave the way for further Bank support to the development of flnancial lntermedlatlon. Finally, the cadastre would become the basis for land-informatlon systems (LIS's) that could be used to improve water resource and envlronment management. In addltion, Bank  - 16 - experience in other countries provides a comparative advantage for ensuring a well-designed approach to establishing a functional cadastre in Algeria. C. Prolect Ob lctives and Summarv Descrigtion 3.3 By supporting the implementation of the cadastre, the project would help the Government create a legal and instltutional framework that would ensure the functioning of a land market and foster the development of private investment in agriculture, housing, and industry. Project objectives are to: (a) strengthen the institutions that implement ti.e cadastre; (b) support the gradual increase of cadastre-related activities over a four-year period; and (c) develop information systems based on cadastre for taxation and other land-related uses.- 3.4 The project would include the following components: (a) a four-year program for the production of cadastral maps and the adjudication of parcels for 3 million ha of rural land and 36,000 ha of urban land in which urban mapping techniques would be tested and computerization of parcel registration and map production introduced; (b) strengthening of ANC; and (c) provision of technical assistance, training, and studies. D. Cadastral Mavpina and Parcel Adiudication 3.5 This component would support the cadastre's increase in production capacity, and would improve the techniques for urban mapping, map drawing, and land registration. Cadastral mapping methods for rural areas have proved cost-effective, but their implementation is still too slow and would be accelerated through improvement of equipment and training. In urban areas, experience is too short to generalize any technique; various techniques would be tested during the first year of the project. Some computer-aided design would be introduced as a part of this component in order to increase the efficiency of the draftsmen, simplify the maintenance of the maps, and prepare for future use of cadastral maps as background for LIS's. For parcel registration, the use of computerized data bases would be promoted to improve the efficiency of the Land Registry and ease the access to the register. 3.6 The new institutional setting of cadastre (para. 2.33) implementation is critical for the success of this component. ANC was recently given full responsibility for cadastral mapping and for title adjudication preparation. 3.7 Cadastre productive capacities would be increased over a four-year period, during which 3 million ha of rural land and 36,000 ha of urban land  - 17 - would be mapped, and 1 milllon parcels adjudLcated. The component includes: (a) launching an aerial photography p:ogram and increasing the productive capacity of INC; (b) surveying four pilot urban zones and developing stereoplotting for urban cadastral mapping; (c) lncreaslng the productive capaclty of ANC ground survey teams and CF offlces; and (d) promotlng the use of computers for land registration and map drawing. Launching an aerial photography program 3.8 INC would manage this aubcomponent wlth the exceptlon of stereoplotting, which ANC would do. Some 18,000 ha of urban areas would be photographed at scale 1:2,000, 1:4,000 or 1:8,000 as well as 1.3 milllon ha of rural areas at scale l:10,000 and 1:20,000. This program would be contracted out by INC through ICB. 3.9 During the same period, INC would acqulre a new aircraft and photographic equipment. This would enable it to implement the second part of the aerlal photography program, which would include a small-scale survey for rural areas (1,000,000 ha at scale 1:20,000) and a larger-scale survey for urban areas (30,000 ha at scale 1:4,000). 3.10 The existing geodetic network ia satLsfactory in the northern part of the terrltory and does not need to be densified in rural areas. In urban areas, lt would be complemented by mlnor controls made by ANC teams and, ln the southern cities, by SPS controls. Thls component would finance the purchase by ANC of three GPS receivers. Stereo preparation would be carried out by ANC teams. 3.11 Orthophotographic maps would be produced by INC for the rural part of the program and for 60 percent of the urban part. The remaining 40 percent of the urban part would be processed by stereoplotting done by ANC. This component would finance the purchase of seven stereoplotters. Surveying four pilot urban zones 3.12 Various mapping techniques would be tested in diverse urban areas to define suitable techniques for contrasting types of urban fabric: dense ancient town centers, single-famlly houslng neighborhoods, and emerging suburbs at the border of the rural belt. Four sites have been defined: Mostaganem (ancient town center), downtown Oran (modern urban fabric), Birtouta ln the Algiers greenbelt, and El Oued (densely populated town in a southern oasis). 3.13 The mapping techniques that would be tested are: (a) dlrect ground survey at scale ls500 in dense zones; (b) stereoplottlng at scale 1:lO00 in moderately dense zones; and (c) orthophotographic maps at scale 1:2000 in less dense zones; in these zones stereoplotting would also be tested. Aerial photographs would be contracted out by INC (para. 3.8), which would also prepare the orthophotographic maps. Stereoplotting would be contracted out by ANC, which would do stereo preparation, aerotrlangulation, and ground survey.  Table 1t Pilot zones Pilot 5one Parcel density area Mapping unit/ha (ha) technique --------------------------------------------------------------__---- Mostaganem (Tidjit) 100 S ground survey Oran 1S 40 stereoplotting Birtouta S 100 stereoplotting orthophoto map El Oued 50 10 stereoplottlng