Report No. 6734-RW Rwanda The Role of the Communes in Socio-Economic Developrnent October 26,1987 South Central and Indian Ocean Department Africa Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipient: only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS Currency Unit Rwanda Franc (RwF) Exchange Rates (annual average) From January 1974 through August 1983 the Rwanda Franc was pegged to the U.S. dollar at a rate of exchange of US$1.00 = RwF 92.8. Since September 1983 the Rwanda Fran_ has been pegged to the SDR at a rate of exchange of SDR 1.00 = RwF 102.71. Year us $1 1980 RwF 92.84 1981 RwF 92.84 1982 RwF 92.84 1983 RwF 94.34 1984 RwF 100.17 1985 RwF 101.26 1986 RwF 87.64 1987 (8 mo.) RwF 80.67 Fiscal Year January 1 - December 31 Standards: metric FOR OMCIL USE ONLY PREFACE An economic memorandum titled 'Rwanda Recent Economic Developments and Current Policy Issues" (Report No. 6191-RW) was issued on October 31, 1986. This report which complements the above memorandum, reviews the role of the communes in Rwands's development and assesses the effective devolution of responsibility for development planning and implementation to the communes. It is based on the findings of a mission that visited Rwanda in September 1986. The members of the mission were: Zeynep Taymas (mission chief), Ellen Bolenge (loan officer), Benoit Millot (education economist), Colette Mukangiliye (general economist), Michel Barnaud (consultant, agricultural economist), Robert Appart 'consultant, rural engineer), Guy Carrin (consultant, health economist), Andre Guichaoua (consultant, sociologist), and Daniel Niro (consultant, local finance expert). The original draft of this report was prepared in French. Ms. Camille Darmon was responsible for processing the various drafts. The report was discussed with the Government in July 1987. This document has a restricted distribution and may be used by recipients only in the perfomance of their official duties Its contents may not otherwise be disclosed without World Bank authorization. TABLE OF CONTENTS Page LIST OF ACRONYMS.......................................... ...... i EXECUTIVE SUMMARY ............................................. iv CHAPTER I. INTRODUCTION ................ ..... 1 PART ONE - THE COMMUNE AS THE FRAMEWORK FOR DEVELOPMENT.... ... 4 CHAPTER Il. THE INSTITUTIONAL FRAMEWORK ... o ........ ............ 4 A. Commune Institutions .... 4 (a) Historical evolution . o. . o. .o. ...............o . ooo 4 (b) The current framework . ... .... .... ....... .. 7 B. Functions of the Commune .... 10 (a) Commune development . . . ...... . .... . 10 (b) Project implementation .. ...o........... . ..... ....... 12 (c) Management of commune assets *..............*...... 13 CHAPTER III. THE FINANCIAL FRAMEWORK .....o ............ oooo.o... 15 Ao Commune Taxes ........................................ 16 (a) General taxes o..o.........................*......0.0 17 (b) Local duties ....17 (c) Special levies ....00#00000***$*0*0*18 (d) Other revenues ooo00#000*0.$00*00* 18 (e) Toward a land tax 19 B. Communal Budgets 20 (a) Current budget .......... o ............................ 20 (b) Investment budget .. ....... .o......... . ... ..... . 20 C. Communal Development Fund ........................... 22 (a) FDC resources .oo ..... ....... ............. ..* 22 (b) FDC operations ... ...... ....... "... 23 (c) The FDC and development finance 24 - 2 - CHAPTER IV. THE SOCIOECONOMIC FRAMEWORK ................ ....... 25 A. The Commune in Relationship to Other Parties 25 (a) Relationship with the Central Government 25 (b) Relationship with communal employees .....26 (c) Relationship with the population .....26 (d) Relationship with local groups 28 B. The Commune as an Economic and Social Reality 29 PART TWO - THE COMMUNE AS CENTER OF ACTIVITIES ............... 33 CHAPTER V. AGRICULTURAL ACTIVITIES ............................... 33 A. The Actors 33 (a) Communal personnel 33 (b) Projects ....... 34 (c) Agricultural cooperatives ....35 (d) The role of women ......................**.. .. 36 B. The Operations 3........37 (a) National campaigns (encadrement) ..................... 38 (b) Extension system ............. .. .. .. .. .. .. .. .......... 39 (c) Support services .......... .. ..*.. .. .. .. .. .. .. .. .. .. . . 42 (d) Land tenure and land use planning ...... ............... 42 CHAPTER VI. SOCIAT. SECTORS . * * * * * * * ... ........ 45 A. Education 5 .......45 (a) Types of education 45 (b) Administration of education in the Communes 46 (c) Financing at commune level ........................... 48 (d) The CERAIs ("Centres d'enseignement rural et artisanal") 51 (e) The CCDFPs ("Centres communaux de doveloppement")..... 52 (f) Private secondary schools 53 B. Public Health ................................ 54 (a) Financing .........55 (b) Cost recovery ..... 56 (c) Pharmaceutical products 57 - 3 - CHAPTER VII. UMUGANDA .... .................. 59 A. Organization of Umuganda 59 B. The Work and the community........... 63 C. Evaluation ....... -..................65 (a) Quantitative results 65 (b) Prospects ..................... 67 CHAPTER VIII. NON-GOVERNMENTAL ORGANIZATIONS .................. 70 A. Multiplicity of Organizations 70 B. Their Activities .. .... ........ 72 C. Their Role ...........76 ANNEXES I. List of Communes visited. ... 78 11. Communal Finance Tables 79 III. Bibliography .......................................................... 83 List of AcronMs ACOR Association de Coordination des ONG pour le Developpement Rural (Association for Coordination of NGOs for Rural Development) ADECOK Le Developpement Communautaire do Kanage (Association for Community Development In Kanage) ADRI Action pour le Developpement Rural Integre (Action for Integrated Rural Development) AFVP Association Fran$aise des Volontaires du Progres (French Association of Volunteers for Progress) AIV Aide Infirmier Veterinaire (Veterinary Assistant) ASBL Association sans But Lucratif (Nonprofit Association) BED Bureau Episcopal de Developpement (Episcopal Office for Development) BGM Bugesera - Gisaka - Migongo BUFMAR Bureau des Formations MNdicales Agre;s du Rwanda (Bureau of Recognized Medical Institutions in Rwanda) CCE Conseil de Commune pour l'Enseignement (Commune Council for Education) CCCE Caisse Centrale de Cooperation Economique, France CCDFP Centre Communal de Developpemert et de Formation Permanente (Commune Center for Development and Continuing Education) CERAI Centre d'Enseignement Rural et Artisanal Integre (Integrated Rural and Artisanal Instruction Center) COFORWA Compagnons Fontainiers Rwandais (Journeymen Water attendants of Rwanda) COOPIBO Compagnons Batissours (Journeymen Builders) CPM Contribution Personnelle Minimum (Minimum Personal Tax) CPR Conseil Protestant du Rwanda (Protestant Council of Rwanda CSA Centre de Sante Agre; (Recognised Health Center) - Li - CSC Centre des Services Cooperatifs (Center for Cooperative Services) CSE Centre de Sante d'Etat (Government Health Center) DGE Direction Generale de l'Eau (General Directorate of Water) EA-A Euro Action-Accord FAC Ponds d'Aide et de Cooperation, France (Aid and Cooperation Fund) FDC Ponds de Developpement Communal (Commune Development Fund) INADES Institut National pour 1e Developpement Economique et Social (National Institute for Economic and Social Development) ISAR Institut des Sciences Agronomiques du Rwanda (Institute of Agricultural Sciences) ITARA Information et Formation en Technologie Appropriee Rurale et Artisanale (Information and Training in Appropriate Rural and Artisanal Technology) IWACU Centre de Fromation et de Recherche Cooperative du Rwanda (Center for Training and Cooperative Research in Rwanda) JOC Jeunesse Ouvriire Catholique (Young Catholic Workers) KORA Artisanal Federations grcuped under the name "To Work' LPR Laboratoire Pharmaceutique du Rwanda (Pharmaceutical Laboratory of Rwanda) MINAGRI Ministere de l'Agriculture et de l'Elevage (Ministry of Agriculture and Livestock) MINEPRISEC Ministere de l'Enseignement Primaire et Secondaire (Ministry of Primary and Secondary Education) MINIJEUCOOP Ministere de la Jeunesse et du Mouvement Cooperatif (Ministry of Youth and the Cooperative Movement) MININTER Ministere de l'Interieur et du Developpement Communal (Ministry of the Interior and >;z.munal Development) MINIPLAN Ministere du Plan (Ministry of Planning) MINISAPASO Ministere de la Sante Publique et des Affaires Sociales (Ministry of Public Health & Social Affairs) MRND Mouvement Revolutionnaire National pur le Developpement (Revolutionary National Movement for Development) - {i - NOO Non-governmental Organization OBM Office du Bugesera-Mayaga OCSD Organisation Canadienne pour la Solidarite et le Developpement (Canadian Organization for Solidarity and Development) ONAPO Office National de la Population (National Population Bureau) OPHAR Office Pharmaceutique du Rwanda (Pharmaceutical Bureau of Rwanda) OVPR Organisation de Volontaires et des Partenaires Rwandais (Organization of Volunteers and Rwandese Partners) PAC Plan d'Action Communal (Communal Action Plan) PCH Direction Generale des Ponts et Chaussees (General Directorate of Public Works) PEV Programme Elargi de Vaccination (Expanded Program for Immunization) SNEC Le Secretariat National de l'Enseignement Catholique (National Secretariat for Catholic Education) SNV Systeme National de Vulgarisation (National Extension System) SVN Service Neerlandais des Volontaires (Dutch Volunteer Service) UNCDF United Nations Capital D.velopment Fund UNDP United Nations Development Program UNIFEM United Nations Development Fund for Women USAID U.S. Agency for International Development - iv - EXECUTIVE SUMMARY i. The "Commune" has been considered for many years by the Government of Rwanda as the "basic development unit." Indeed, from an administrative viewpoint, it is the local authority closest to both the rural and the urban population. Over time, the commune has become a focal point not only of political and administrative but also of economic and financial activities, and hence an essential element in the country's economic-and social development strategy. In order to better understand this strategy, the WoULd Ba1k, in collaboration with the Rwandese administration, has undertaken this introductory study. The major conclusions and recom- mendations of the study are presented here in a condensed form for the readers who are familiar with the communal organization in Rwanda. A first set of conclusions are general and relate to the overall policy of communal development. The other conclusions are more specific, dealing with various aspects of a particular institution or a special policy. ii. The most general conclusion is that the communes have proven to be viable and on the whole effective entities in service of the nation's political, social and economic objectives. With regard to the implementation of their economic objectives, which is the main focus of this report, the zommunes have achieved notable success in mobilizing substantial amounts of human and financial resources and using them at the local level to provide essential services as well as to raise productive capacities through investments in social and economic infrastructure. The benefits of these activities have been reflected in improved living standards. Thus, this report's diagnosis of the communes' roles in the developmental process reveals considerable strengths and successes that can and should be built upon. iii. The the report has also identified certain shortcomings and constraints which need to be overcome in order for the communes to strengthen the effectiveness of their resource mobilization and to improve the efficiency of their resource allocation and utilization. Taxation in particular has provided the communes with a notable measure of fiscal autonomy, although in some cases its proceeds have barely sufficed to pay for core services. Local officials have been most successful in establishing a relatively efficient organization and in carrying out their administrative duties. Yet the communes still fall short of their potential as the "basic development unit", not only because of tbeir insufficient financial and manpower resources, but also because the Central Government has adopted an excessively rigid and centralized approach to specifying the multiple tasks to be executed by the communes. Hence the communal administrations are often unable to reconcile generally legitimate government concerns with local conditions and farmers' constraints. They are also deficient in technical expertise needed for project and program planning and implementation at the local level. iv. Three general recommendations are thus presented: (i) a decentralization of some central authority in favor of the communes; (ii) a scaling down of some powers of the commune while encouraging citizens' - v - initiatives; (Uii) a fuller accounting of the diversity in local conditions to be made by authorities at all levels in the specification of tasks. An additional recommendation is the strengthening of training programs for elected officials and technical personnel employed by the communes. The first general theme is a reexamination of the way authority is distributed among the institutions external to the communes (the Central Government, the Party, the Communal Development Fund (FDC), projects, etc.) and the communes themselves. The relationship with the central authorities should be based on Government's encouragement of communal initiatives. National oblectives should be promoted through incentives and contractual agreements rather than hierarchical instructions. It would be more beneficial if the Government were to limit its interventions in commune management and not impose unqualified quantitative targets to them. At the same time, the communes should have a more extensive participation in the management of the FDC. There is also a need to develop the necessary institutional arrangements for intercommunal cooperation to implement specific projects concerning several communes (for water supply, forestry, municipal services, etc.). Externally-financed and managed projects should also give greater consideration to the authority, institutions and competence of the communes by supporting existing institutions and group initiatives. v. The analysis suggests that the burgomaster has too many responsabilities, even if he performs them well in general. It is, therefore, necessary to decentralize administrative power within the commune. The bureomaster should be in a position to delegate authority to other officials in the commune. This is less a problem of organization -- which is satisfactory on the whole -- than of motivation and training. It mav be adv-'sable to provide some training through the CCDFPs (see para.xx) to communal officials in the "secteurs" and "cellules". vi. The design and preparation of individual communal programs could provide an opportunity for a greater decentralization. The multiplicity of the tasks assigned to and undertaken by the communes strongly suggest a need to set priorities on a multi-year basis, and to plan future investments, the work program of Umuxanda, and mobilization of resources (local and external) accordingly. Indeed, while development goals can be set for the country as a whole, methods for achieving them sheuld be worked out by the communes in consultation with the population. vii. The second general theme is the need to reduce the extension of central and local go-.rnement by taking greater account of the population at lurge, local groups, and other private initiatives. The Rwandese Government has largely succeeded in establishing an effective system of communal organization; it should now cease expecting the communes to assume responsibility for all grassroots development activities. Certain tasks could be transferred to economic agents other than those of the administration. Although the interventions of the commune are totallv justified in the area of health and hygiene, their involvement in productive and commercial activities should be limited to the promotion of the private sector. Throughout Rwanda, individual initiative is very prevalent; a good example is the establishment of private secondary s.naools. In the productive and commercial sectors, however, communal activities sometimes encroach on areas that could more appropriately be left to the private sector. Burgomasters have only minimal direct interest in promoting economic growth in general, as they are too absorbed in - vi - carrying out administrative duties, mobilizing communal resources and executing government campaigns. In order to meet government dmands, the communes often frustrate private initiative. They may undertake remunerative activities to increase budgetary revenues although such operation is in contradiction with the Government's explicit policy. The Government should review the role of the communes on a more systematic and well-informed basis (possibly in the context of the Fourth Plan preparation) than has been possible within the scope of this report. This critical review may lead in a number of instances to streamlining communal assisgnments and, more important, may result in a different approach to the administrative hierarchy as a whole and its interactions with the population. viii. The widespread presence of cooperative, associative and risk- tharins groups is largely responsible for the vitality of local communities. First, cooreratives have official status and receive government su?port through the commune. Second, there are numerous other farmers' groups and "tontines" more or less formally organized and subject to very little control by the communal authorities. Little is known about these groups. More extensive knowledge of these groups would be necessary -- not to control them but to help them benefit from the modern economic system. The communal administration should encourage and assist the local Sroups to formalize" gradually and to obtcin easier access to services such as extension and bank credit. ix. More importance should be given to the role of women in development. They not only account for an extremely active half of the population, but also afford a potential for initiative, of which the country should be able to take advantage. They constitute an important element in the country's strategy of food self-sufficiency since it is mostly women who are in charge of growing fooderops. The Government has undeniably made efforts to improve their situation and has achieved significant results in education and health. Nevertheless, this report gives several examples of bias at the commune level against spontaneous all-female groups. Specific areas for improvement would include the development of appropriate technologfy (water tanks, grain mills, pulping equipment, grain silos, etc.), n increase in the number of female extension agents, a better participation in CCDFP training, and a greater support to women's associations. x. The third general theme is the need to bring the activities of comnal institutions more in line vith local ct)aditions. The diversity found among regions, communes, and individuals could then be more fully reflected in development programs. The primary task of the Government after Independence was to set up political and administrative institutions that would unify the country. This objective has by now been largely completed, and the administr%tive framework is reasonably efficient. At present, it would be preferable to give the diversity observed in Rwanda free play within what is an already firmly established setting. For instance, communal programs and interventions could be designed with greater care in agriculture and Umuganda. In these ca8es, considerable organizational and human energy has been mobilized for justifiable ends but with results that are at times hardly satisfactory. The govenment campaigns, the agricultural extension messages, and the community work are too often the product of centralized decisions which take very little - vii - account of realities. It is clear, all the same, that such shortcomings are not only characteristic of the power exercised by the central authorities; they also reflect weaknesses in the transmission and execution of instructions and a lack of suitable training among executive and technical personiel at all levels. This, however, can be remedied only over the long term, and the Government is already taking steps to do so. In the interim, means are available -- and the report enumerates some of them -- for channeling these energies in a more efficient way by adapting them better to the country's ecological, agricultural and economic conditions. xi. More specific recommendations are given in the report. They relate to the following areas: (i) communal finances, (ii) agriculture, (iii) education, (iv) health care, (v) Umuganda, and (vi) non-governmental organizatioas. As regards comnunal finances, the framework is well- established. Without major recourse to the Central Government,l the communes have mobilized substantial financial resources, even though these are often inadequate to meet the vast array of tasks that have devolved upon them. The total current revenues of the 143 communes were equivalent to approximately 6 percent of Central Government's tax revenues in 1985. The burden of communal taxes is estimated at not less than 5 percent per annual family monetary income, urban and rural alike. xii. Communal budget revenues are numerous. First, capitation (CPM) and livestock taxes are still being levied in Rwanda. Second, local duties on economic activities are the category of communal taxes that has risen most sharply in recent years. These taxes should not be overused, however, as they can ultimately reduce the tax basis by discouraging productive activities. It is recommended that a study of local duties be undertaken to analyze the impact of these taxes on economic activities. Finallyg in addition to other current revenues and receipts, most communes have introduced -- not totally in line with current legislation -- special levies (known as "umusanzu") that are imposed ad hoc to finance specific investments.2 In a modern fiscal system head taxes should be integrated into the national income tax. This can, however, be achieved only over the long run as Rwanda's rural economy becomes more monetized. In the interim, the rate of CPM could be revised, include the umusanzu and be graduated according to the degree of wealth and monetarisation of communes and/or regions. xiii. In the area of local taxation, the report reconmiends the gradual introduction of a land tax preceeded by cadastral surveys. Currently, the Central Government levies two taxes on real property. Their proceeds are very limited and they take no account of the location, the use or the quality of the property. The existing local taxes may be replaced in time by the proposed land tax, the proceeds of which might be handed over, in whole or in part, to the communes. This system of land taxation should be 1/In a sample of 18 commune accounts, government subsidies represented 4 percent of total revenue for the period 1980-85. 2/These "umusanzu" have been subject to so much criticism from the population that a ministerial decree was promulgated in order to limit the recourse to this taxation. - viii - put in place first in urban communes after a pilot operation within the IDA-assisted Kigali Urban Project. xiv. The communes' investments are equivalent to about half Lheir current expenditures and roughly 17 percent of the Central Government's development budget. The investment efforts by the communes are more important than indicated in their accounts, as these do not include investments by NGOs, foreign aid agencies, and the works undertaken through Umuganda. The proportion of the investment budget that is self-financed is not insignificant: in the sample of 18 communes, 30 percent of investment in 1980-85 were financed out of "umusansu'. Some communes did not resort to this form of taxation at all, while others funded more than 70 percent of their investment budgets from such compulsory levies. Part of communal investments are financed by the Communal Development Fund (FDC) which was established to effect some equalization among communes and to give them access to credit, either in the form of direct loans or by guaranteeing their borrowing from banks3 . xv. FDC's administrative capabilities seem to be limited with respect to:t (i) appraisal and follow-up of projects; (ii) the accounting records; and (iii) forward planning. A far-reaching reform of FDC is thus needed to better organize the intermediation of centrally provided resources. To become en effective institution for the financing of communal development, FDC should: (i) recover its separate legal status and financial autonomy under normal Government supervision; (ii) strengthen its management; (iii) review its project selection criteria and its procedures for allocating its own funds to take better account not only of the poverty and needs of each commune but also of its relative efficiency andlor its creditworthiness; and (iv) finance more relevant projects of economic and social infrastructure as proposed by the communes themselves. To that effect, financial resources can be intermediated by the FDC to the communes through specialized 'windows.' Such an approach (as proposed in the water supply program) can be used to support rural road improvements, maintenance/ construction of primary schools and other infrastructure within the management capacity of the communes. In this regard, the technical capabilities of the communes to prepare projects also need to be improved, since they have somewhat inhibited FDC's operations. xvi. The major activities of the commune in agriculture consist of: (i) execution of government campaigns ("encadrement"); (ii) extension of technical messages; (iii) provision of farm services; and (iv) management of the land holdings. Many such campaigns, anti-erosion or reforestation in particular, have brought significant benefits. In other cases, the somewhat bureaucratic approach used has had adverse effects as well as a high cost in the use of human resources (time wasted and loss of enthusiasm among both campaign and extension personnel and the farmers themselves). A clearer distinction should be made (in organization and implementation) between extension work and government campaign activitiess the communes should enter into contractual agreements with MINAGRI for the implementation of campaign activities. These can be carried out by commune 3/FDC's own funds come from annual contributions of the communes (10 percent of their current revenues); these have provided close to 40 percent of FDC's resources since its inception. - ix - councilors (after appropriate training), instead of agronomists who should focus solely on extermion activities. In the provision of services. the communes should restrict their inwolvement to those of public interest such as meat inspection, production of coffee and other tree seedlings and anti- erosion and pasture plants. xvii. Land use and tenure policy needs to be better adapted to present conditions -- by improving land development, limiting certain forms of land accumulation, increasing communal revenues and facilitating provision of agricultural credit. To this end, land legislation should be actively reviewed following a thorough study. In the interim, a gradual rationalization of land use patterns is necessary. The commune's role in this process may consist of: (i) undertaking of cadastral surveys which would provide better information on spatial land distribution; and (ii) establishment of "area development contracts" between farming communities with the possible assistance of a national NGO. Management of communal land and forests should be handled within a stricter accounting framework of a commercial type; in specific cases, it might be advisable to enter into management contracts (private companies or !nter-communal forestry management partnerships) within the rules and regulations of the new Forestry Law. xviii. In education the commune plays a crucial role. At the communal level, the education system is characterized by: (i) organizational complexity, since every individual involved in the educational system is represented on one of the numerous committes; (ii) engagement of many actors in the collection and management of financial resources, with the risk of lower efficiency and accountability; (iii) financial constraints, despite substantial ad hoc contributions of the population; (iv) presence of public institutions that supplement the conventional education system; and (v) mushrooming of private secondary schools. xix. The commune authorities contribute 6 percent of the total financing required in education at the commune level (i.e., approximately 10 percent of the commune budget) besides contributions in kind. The local population (as both residents and parents of pupils) covers 20 percent of the total costs through payment of the "umusanzu" and the "minerval" (school fees) -- in addition to their contributions of labor through the Umuganda. The communal authorities play an important role (either directly or indirectly) in deciding how these resources are to be used. It is necessary to define more clearly the respective authority of school management and the communal administration in the allocation of communal funds, the minertal and the umusanzu. xx. The CERAIs and the CCDFPs are purely communal institutions. The CCDFPs, centers of "literacy and non-formal training," provide an instruction with both technical and political content within each commune. A wide range of personnel (mostly paid from the communal current budget) is active in these institutions. The CERAIs, centers of "integrated instruction in rural and occupational skills," were set up to compensate for the shortage of student places in secondary education and to provide the skills necessary to enter the labor force in the commune itself. However, very few CERAI graduates are able to find an appropriate Job locally, and the institutions are now having difficulty in attracting students. The lack of a clearly defined vocation, their uncertain tie-in w!'h a long primary education cycle, their poorly qualified teaching petsonnel, and the lack of ready outlets for their new graduates constitute a set of problems which the communes are hardly In a position to solve. It is advisable to undertake a thorough study on the employment possibilities and prospects of CERAI students with a view to improving CERAI curriculum. xxi. The private secondary schools are usually established by religious groups and parents' associations. The growing and relatively recent initiatives of parents' associations in this tespect reflect a response to a pressing social demand for secondary education which the public sector is unable to provide. The burgomaster frequently serves as a catalyst in the establishment of these facilities, which often receive support from the communes in the form of a loan or an assignment of land. These schools offer mostly low-cost tertiary-stream subjects; technical training is poorly represented. The Central Government should closely monitor the development of private secondary schools and, if needed, provide guidance to these establishments, in the light of overall manpower requirements. xxii. The role assigned to the commune with respect to health care facilities is somewhat unclear, and this uncertainty has occasionally led to functional and financial difficulties, as noted below. It would be advisable to establish more formal and organic contacts between public health centers (CSE) managers and commune authorities, perhaps through the creation at each CSE of a management committee. Participation by the communes itl health care activities as well as in their financing varies considerably from one to another. Many communes complain of the large deficits in their health buigets, despite the recent transfer to communal budgets of the patient fees which were previously transferred by the CSEs to the central government budget. These rates, however, have not changed since 1974. Increased fees would be accepted by the population, especially if accompanied by improved availability of pharmaceutical products. The general shortage of such goods has been partially mitigated by the establishment of a pharmacy in most communes. Nonetheless, national distribution and pricing of pharmaceuticals face difficulties which are beyond the capacity of the communes to handle. A government agency and various NGOs supply the CSEs with less costly generic products, but pharmacies can only obtain brand-name products imported by private wholesalers at high cost. A study of the issues of the distribution of pharmaceutical products is needed, with a view to revising current institutional arrangements in this field. xxiii. Umuganda, the concept of non-remunerated community work, is a national effort undertaken entirely at the commune level. All able-bodied adults are expected to participate in "Umuganda" by contributing one half- day per week, although in practice the actual contribution is much lover. Every burgomaster is responsible for organizing Umuganda in his commune. Like local taxation, Umuganda is one of the burgomaster's key instruments for ensuring the implementation of the tasks for which he is responsible. In many respects, Umuganda reflects his management style: in some communes, the working conditions are relatively flexible, in others 'the main concern is that people obey and learn to work for the community.". When benefits are apparent, the population is eager to obtain the results and well aware that there is no other means of achieving 'hem. Despite the constraints imposed by Umuganda on the population, it provides the opportunity for the - xi - communities to become materially involved in collective projects that affect them directly. xxiv. A strict economic assessment of Umuganda is difficult in quantitative terms. Umuganda may have contributed 5 to 6 percent of the total investment undertaken during the last two Plan periods. Despite the large amount of work completed, its effectiveness is not always evident. Nonetheless, the overall system is carefully organized and accompanied by very active sensitization efforts and fairly systematic monitoring. There is also an on-going process of dialogue at various levels of government on problems encountered in implementing Umuganda. There is room for improvements and adjustments in several areas - institutional, technical, economic, and psychological. First, the communes need to be allowed treater flexibility and scope for innovation so that they can adapt Umuganda better to local conditions. Second, greater emphasis could be placed on technical considerations rather than on priorities that are too exclusively political; for instance, specialized training for workers and procurement of appropriate tools would make it possible to set precise quantified goals. Some tasks could be delegated to users' associations and cooperatives which would organize community work in a more participatory, creative and flexible fashion. Finally, economic incentives could supplement political motivations, for organizers as well as for Participants. Such actions are obviously not in themselves sufficient to accomplish all the objectives of Umuganda, but they would help make it less rigid and constraining. xxv. Non-governmental organizations (NGOs) are one of the most important actors in the social and economic development from below. There are over 160 NGOs in Rwanda, approximately 130 of them active at the commune level. In 1984, their quantitative contribution to the Rwandese economy was equivalent to about 10 percent of national investment and 75 percent of the central government's development budget. The NGOs are active in many fields, particularly in public health, education, training, and water supply. They also provide the communes with technical assistance in preparing productive projects and improving the operations of locally- based groups. Although approximately 80 percent of their work is socially oriented, the NGOs are taking an increasingly large part in directly productive activities in agriculture and lartisanat,u and in assisting the informal urban sector. It is estimated that 70 percent of the NGOs active in Rwanda are working to benefit women. There are also some 50 'twinning" agreements between Rwandese communes and communes in other countries (especially in Germany). xxvi. For the communes, the NGOt are preferred partners in the national effort of economic and social development, since they have demonstrated greater flexibility than the large aid agencies in implementing grassroots projects. All the same, there is room for improvement in certain aspects of NGO activity on both the Government and NGO sides. Specifically, better coordination at government level is necessary not to control NGO activities but to ensure consistency of their actions with national policies. including interregional equity. More consideration should also be given to the economic aspects of NGO intervention. Their projects (which emphasize social components) often do not pay sufficient attention to the economic returns which would ensure project sustainability. It is becoming increasingly important, in a decentralized development strategy, for NGOs - xii - and communal authorities to focus their activities on spontaneous groups and to encourage private sector initiatives. Greater effort still is needed to ensure the success of these initiatives and to determine more accurately how aid agencies can work closely with NGOs without either side losing its own effectiveness. xxvii. In sum, the main conclusions and recommendations of the report are that the communes could become even more effective instruments of development in Rwanda than they have already proven to be if: (i) interventions and directives from the Central Government, as well as the communes' own initiatives, were more selective; (ii) there were a greater decentralization of administration at the commune level (reducing the burden on the burgomaster and strengthening grass roots participation); (iii) emphasis were placed on actions perceived as directly beneficial by the populations themselves; (iv) closer attention were accorded in the specification of tasks to local conditions, both physical and socially, and; (v) efforts were made to strengthen communal-level technical expertise needed for more effective planning and implementation of communal projects and programs. CHAPTER I. INTRODUCTION 1. The 'Commune" has been considered for many years by the Rwandese Government as the "basic development unit." Indeed, from an administrative viewpoint, it is the local authority closest to both the rural and the urban populstion. Over time, the commune has become a focal point not only of political and administrative but also of economic and financial activities, and hence an essential element in the country's economic and social development strategy. In order to better understand this strategy, and to help integrate more adequately external aid (including its own) with local realities, the World Bank has undertaken the present study of the complex problems of economic administration at the commune level. 2. Worldwide, grassroots development has taken various forms since the 1950s: (i) "community development" tried to combine technical expertise of the central government with local participation; (ii) then, "integrated rural development' attempted to promote development on all fronts, i.e., production as well as social infrastructure; and (iii) more recently, 'decentralized development" that encourages initiatives of the various agents at the local level. While the Rwandese approach undoubtedly has similar origins, it has certain specific characteristics in relation to these strategies. 3. It is not the purpose of this Report to try to compare these various experiences or to gauge the results obtained in Rwanda. This is a highly complex undertaking that would require a knowledge of the country beyond what can be achieved within the context of this study. The multiplicity of institutions and of actions involved within the commune is such that analysis must often be limited to understanding how they fit together. A description and understanding of the facts must precede their evaluation; in many respects, therefore, this study is only an introduction.1 4. This study was particularly difficult and delicate. It had to analyze the operations of the Rwandese administration in an overall framework, and to establish the link between the relatively little known institutions at the commune level and a wide range of economic and social actions. Also, the issues raised in the context of the study were relatively sensitive because the commune in Rwanda is at the center of an open political debate which is trying to define future orientations and options for its development. 5. In a country facing formidable national and regional challenges, where traditions of centralism and hyperorganization have centuries-old roots, social and political motivations cannot be isolated from economic and technical considerations. Therefore, this study is not limited to purely sectoral (agriculture, education, health care) and financial (local taxation) aspects but has also attempted to analyze the socio-institutional framework at the local level, including a largely political area, namely "Umuganda". This approach is considered at every echelon in Rwanda as a necessary first step to understand I/There is no overall study of the communes in Rwanda. - 2 - the stakes involved in communal development.2 On the other hand, the study does not delve into some socio-economic issues, such as regional development or urbanization, which can only be analyzed after the institutional and social situation has been more fully understood. 6. The commune is the basic administrative subdivision in Rwanda: the country is divided into 10 prefectures and 143 communes (11 of these being urban communes).3 It is also a territorial entity of variable size, but with a population generally averaging around 40,000 inhabitants. It is also a decentralized unit with legal status, with the ability to enter into contracts, with its own assets and with budgetary autonomy. It is subject to the control of the Central Government, which exercises supervision ("tutelle") over both its acts and its instituti)ns. 7. For the purposes of the mission, a group of 10 communes spread over eight prefectures was selected. This sample was studied by all the members of the mission. Nine additional communes were visited by the various mission members, reflecting their special interests (agricultural development, commune pharmacies, schools set up by parents' associations, etc.). This sample (see list in Annex I) was selected to cover a certain variety of situations, but does not claim to be statistically representative. 8. Within the confines of a three-week mission, one could not conduct in- depth surveys. The information and opinions obtained were collected at four levels.4 First$ discussions with public officials (in ministries, prefectures, and communes) were held in a semi-structured format on issues systematically raised with each person interviewed. These interviews prompted fruitful and well-argued discussions which were generally very direct and frank. To gauge the views of the population at large -- which could not be ascertained in the course of official visits -- the study relied essentially on the reports and findings of various earlier surveys and questionnaires on related topics; these had typically been carried out by the mission members themselves in the context of other studies. The mission also exchanged views with aid donors, including 2/"You cannot properly understand a technical decision unless you know who was pushing it and who profited from it"; "You cannot be an effective government official until you come to realize why it is not always the best technical reports that are accepted," etc. (statements by high government officials). 3/Except in the case of the Prefecture of Butare, where there are two urban communes, the epithet "urban" applies to the chief town in the prefecture, even if in certain cases its level of development or economic situation does not really justify its being placed in that category. At least, one intermediate category should be established for later studies, such as 'market town" (centre de negoce) where certain economic activities of particular relevance are concentrated, as in the case of a market serving a sizable region. 4/There are also reports ("monographies") prepared annually by each of the communes outlining its key geographical, demographic and economic data. non-governmental organizations (NGOs), both in the capital city and in the field. Lastly, the quantitative and qualitative information gathered, representing both fact and opinion, was supplemented by documents made available by the various ministries. 9. This report presents the mission's findings in two parts. First, it tries to asc*rtain to what extent the commune offers an appropriate framework, both institutional and functional, for economic and social development (Chapter II). The commune is also the unit which through financial resource mobilization and the Communal Development Fund (Fonds de Developpement Communal) plays a major role in financing activities at the grass roots (Chapter III). The commune thus emerges as a socio-economic reality enhanced by a strong administrative structure and the mobilization of substantial local financial and human (umuganda) resources. Its future development will depend in large measure on the degree of decentralization allowed and the use it makes of its powers, not only with regard to the Central Government but also vis-a-vis economic actors at the local level (Chapter IV). 10. The second part of the report describes how various economic and social activities have found in the commune the most adequate setting for delivering services effectively. The commune thus appears as a highly complex structure where a large number of development agents conduct a large number of activities. These multiple actions are reflected in a plurality of local institutions fitting within the somewhat rigid but relatively well organized framework provided by the commune. 11. Agricultural development, illustrates the various problems posed when the Central Government, the commune, the local population and external aid agencies are all involved (Chapter V). Social activities of the communes, mainly in the area of education and public health, are of particular interest to the population. The participation of the users and the residents of the commune in financing and managing these activities is quite important. Moreover, where the public sector fails to meet the social demand, private initiatives take over, as in the case of private secondary schools (Chapter VI). Local population contributes directly to communal development by participating in non- remunerated community work ("Umuganda"), whose multiple objectives often fail to take into account efficiency considerations (Chapter VII). The final chapter presents an overview of the operations of non-governmental organizations in the various economic and social sectors of Rwanda's communes. - 4 - PART ONE THE COMMUNE AS THE FRAMEWORK FOR DEVELOPMENT CHAPTER II. THE INSTITUTIONAL FRAMEWORK A. Commune Institutions (a) Historical evolution 12. A brief historical survey of the administrative organization in Rwanda is useful to comprehend the current framework. This will also enable readers to understand the perception that the oldest members of the community -- who still occupy responsible posts or enjoy considerable authority among the population -- may have of the successive changes they have witnessed since the establishment of the colonial administration. While there are no longer many old people on the hilltops "collines" (para. 21) who actually experienced German colonization at first hand, oral traditions are still rich regarding the period of Belgian trusteeship and the coming of Independence. These references, which are largely unknown to the younger generations of urban civil servants, play an essential role in the countryside and are frequently brought up in discussions. 13. The first major territorial reorganization was carried out by the colonial power in 1926, when chiefs loyal to the monarchy and to Mwami Musinga were replaced. Younger chiefs trained in the 'official schools" were given responsibility for new chiefdoms and subehiefdoms. At the time the country was divided into nine territories and 45 chiefdoms split up on a rather artificial basis. 565 subchiefdoms formed the basic units of the traditional sociopolitical structure. The introduction of this organizational set-up and its new officials represented an extension of the royal tradition of 'concentrated centralization".5 These changes immediately preceded the first decisions that have left their mark to the present day on rural life. 14. An authoritarian system of agricultural extension was introduced to enforce measures .elating to erosion control, reforestation, use of organic fertilizer (compost pits), growing of cash crops (coffee) or food security crops (cassava), as well as construction of earth roads, latrines, and water points. These compulsory measures, the generalization of taxes and forced labor, and the punishments that accompanied them, added to the emigration pressures stemming from overpopulation. On the eve of Independence in 1959, there were some 350,000 "emigres" living in Tanganyika. There was also an organized emigration to the agricultural regions of Kivu and mines in Shaba, in Belgian Congo. 15. This forced introduction of agricultural "innovations" marked the population and largely explains why certain measures (nurseries, selected seed, breeding stations, tick baths, etc.) were not more widely accepted by the rural 5lIn a "centralized" state organization, representatives of the Central Government manage local affairs. Power is 'concentrated" if these representatives cannot take any decisions without reference to the hierarchically higher level. masses even though their short-term effects on output and on incomes was probably evident to the farmers. With the coming of Independence, accompanied by widespread social disturbances in the countryside, most of these practices (erosion control measures, compulsory labor, etc.) were regarded as symbols of the colonial era and rejected on that score. It took several years before the Rwandese authorities responsible for agricultural development established a certain extension capacity and began to use their powers in this area. 16. An administrative and political reorganization of the country had taken place in the years prior to Independence in 1962. In 1952, a system of popular representation was established (with subchiefdom councils selected by and from an electoral college of local notables). This presaged a process which led in 1960 to the election of the first commune councilors and burgomasters (paras. 21-23). As the communes were headed by councils elected by popular suffrage, they were considered by the population as the sole representative bodies. This was so much the case that one scholar felt entitled to write: "In the course of a single day, the burgomasters transformed Rwanda from a monarchy into a republic, elected a legislative assembly and the first president of the new republic, and also adopted a provisional constitution."6 .The Organic Law of 19637 also introduced the principle of communal autonomy in financial and administrative affairs.8 17. As in many African countries, the period of accession to Independence in 1962 was reflected in considerable strengthening of local communities. Local authorities exercised power for the first time in history, and -- whether officially sanctioned or not -- started running local affairs themselves. The people became accustomed to living and working independently of the central authorities. The power vacuum at the center often left room for members of the local elite.This experience gave the local communities a solid footing in society which survived later on, even when they were again subjected to stricter control. 6/A. Bosomingera, 'La participation de la popufation a l'administration et l"execution des missions de developpement des collectivites locales au Rwanda," in Les Collectivites Locales et le Pouvoir Central, Editions Nathan, Paris, 1984. 7/Law on Communal Organization (23 November 1963). 8/It must, however, be noted that, while recognition of the legal status of the commune is a condition for decentralization, it is not enough to guarantee it. In practice, the use of legal arguments to defend oneself against pressure from the central authorities is largely illusory, particularly when the executive branch is composed of officials appointed by the Head of State. Rare indeed are the cases where central authority has been defied. The best- known examples are the village ujamaa in Tanzania, the fokonolona in Madagascar and the rural and urban associations in Ethiopia; but here, too, it was not long before they were taken firmly in hand by the central authorities. -6- CHART I ADMINISTRATIVE ORGANIZATION Minstrie MIAGRI INISAPA. MININTER MIPLAN MINIPRIS ETC. - I (assemblies) (chairman) PREFECTURE Council Prefect e I _ (appointed) (SUB-PREFECTURE) Sub-Prefect I (appointed) COMMUNE Council Burgomaster (elected) (appointed) SECTEUR Committee Councilor I I 0 (elected) (elected) |CELLULE Responsable (elected) - 7 - 18. The proclamation of the Second Republic in 1973 brought - strong presidential regime and a marked return to a centralization of powers. The major changes introduced at this time includedt (i) abolishment of the election of burgomasters, who were to be appointed by the President;9 (ii) a blurring of the administrative and political systems with the creation of the sole party, the "Mouvement Revolutionnaire National pour le Developpement" (MRND), the "functional duplication" of certain local government bodies and Party bodies, and the alignment between administrative subdivisions and those of the Party; and (iii) strengthening of the powers of the prefectures (endowed with legal personality in 1975) and the establishment of subprefectures (para. 21). 19. In a parallel move to this loss of political power, the role of the communes as "agents of development" was ttrengthenod. While the commune was becoming merely a Ideconcentrated' element in the State apparatus for administrative purposes, it took on, de facto, a "decentralized" existence through local taxation, the restoration of community labor, and the implementat;.on of communal projects. 20. The importance henceforth accorded to "communal development" was recognized ever more decisively by the national authorities. The priority task was to establish "peace and unity among all citizens". Starting in 1975, the focus on communes took a concrete form with the drafting of a doctrine and lines of action (para. 31). Within the framework of the Third Plan (1981-86), in 1984 a draft second communal program was prepared by MINIPLAN (para. 31). (b) The current framework 21. The national territory is divided into ten prefectures encompassing 143 communes with .egal personality and financial autonomy. The commune represents the lowest administrative echelon in the Rwandese Republic (Chart I). Twenty-two subprefectu:es have been set up to serve the communes farther away from the prefecture. The commune is divided into about ten "Secteurs" (sectors), each headed by a "Con-siller Communal" (Commune Councilor). Each "secteur" is, in turn, divided into "Cellules" (units or cells), representing groups of "collines" (hilltop communities) headed by a "Responsable" (Leader). 22. The Mouvement Revolutionnaire National pour le Developpement (MLND), the political organization grouping all Rwandese citizens, is represented at every level in the comaune by its own institutions: commune congress and committee, "secteur" tssembly and committee, and "cellule" assembly and committee. Communal action is the responsibility of the MRND, which "formulates the philosophy," and the Ministry of the Interior and Communal Development (MININTER), which "has responsibility for implementation." 9/"What was needed above all was to increase the efficiency of the burgomasters, who spent their time flattering the people instead of making them work, in order to guarantee their reelections," (MININTER document). - 8 - 23. The "bourimestre" (burgomaster), who is appointed by the President on the recommendation of the Minister of the Interior, heads the communal institutions. As the representative of the central authority and the official responsible for application and implementation of national laws, decrees, ordinances and recommendations (including those of the technical ministries), he exerts great authority. Statements such as, "The burgomaster, is the commune," or "All depends on who the burgomaster is", are commonly heard in Rwanda. 24. The burgomaster's main concern is to collect revenues for the communal budget, not only to pay its employees, but also to carry out the vast array of tasks assigned to its adw{nistration. The direct taxes assigned to the commune by the Central Government have limited yields; thus the commune is led to impose and collect many different levies and other charges (Chapter III). Another area, where the burgomaster describes his duty as the exercise of a "necessary pressure," is the implementation of community labor (Uruganda) (Chapter VII). Of particular importance is also the organization of political activities, to which the population is expected to devote as much time as to Umuganda.10 Lastly, the burgomaster also plays, a conciliatory role in local disputes (mostly about land); this last role is particularly important, as it determines his reputation among the people he governs. 25. The population at large is represented by the Commune Council ("Conseil Communal") (see Chart II), which "oversees the development of the commune." The Council is made up of as many elected members (commune councilors) as the commune has "secteurs"; the burgomaster chairs its meetings. The actions of the communes are monitored by the Minister of the Interior and by the Prefects (and, '- delegation, the Subprefects). The Council often serves merely as a link in the chain of command for implementation of decisions by higher authorities and the burgomaster. Depending on the commune, the personal prestige of the burgomaster and his seniority, the selection and final choice of candidates for public elections will be largely subject to the burgomaster's approval. 26. The preeminence of the burgomaster is even more apparent in the two consultative bodies within the commune: the Technical Commission ("Commission Technique") and the Development Council ("Conseil de Developpement"). The Technical Commission's task is to help prepare development projects and seek their financing. It comprises five or six members selected by the burgomaster for their "skills in the economic, social, cultural and technical areas". It is an institution dependent on the burgomaster, who selects its members from among leading local figures (not less than one third) and senior government officials and other important personalities (originating from the commune). The composition of this commission (ministers, high level officials, members of the National Assembly, clergymen, etc.) plays a key role in the image the commune projects to the outside world -- and thus in the commune's ability to attract external financing and carry out projects. However, this arrangement also contribvtes, perhaps involuntarily, to stripping the commune councilors of a 10/"The burgomaster's essential duty is to "inspire" the local population and officials to implement the ideals of the MRND" (underlined in the MININTER document). -9- CHART II COMMUNAL ADMINISTRATION < ~~~~~OMN oCOUNCIL TECHNICAL DEVELOPMENT COUNCIL COUNCIL (Examples of sectors) Agriculture Education Health Water Supply Umuganda I N S Extension Communal CSE Users' Umuganda T Committee Council for Association Commission I Education T CCDFP CSA NGOs U I T Public CERAI Hospital I Enterprises CCDFP 0 N NGOs NGOs NGOs S - 10 - large share of their powers, all in the name of efficiency, competence and personal relations. In many communes, the role played by the Technical Commission is largely formal, as it meets infrequently and at irregular intervals. Its very existence, however, gives the burgomaster a valuable support for attracting development projects and resources to the commune. 27. The Development Council is the body serving as "liaison between the commune and the government authorities responsible for planning." Under the chairmanship of the burgomaster it is made up of commune councilors, members of the Technical Commission and officials assigned to the commune. The Development Council is responsible for appraising and monitoring projects, and ensuring coordination among the various parties involved in communal development; but, usually it only expresses an ex post opinion on project documents that have already been agreed upon. 28. The burgomasters very frequently complain of "having to do everything themselves," of "being without competent assistants," of "not being able to rely on the efforts of the members of the Technical Commission, since they are concerned primarily with their own affairs," or of "opposition from the commune councilors or cell leaders (responsables de cellules)." The problem is a real one, and is forcibly expressed in demands for higher financial compensation (for members of the Technical Commission) or for payment proportional to the amount of work expected from other elected officials of the communal administration. The burgomaster himself has similar demands but is obliged to remain in post, unlike the other officials who can stop working, quit attending meetings, or refuse to stand for re-election. It would appear that several communes have vacancies for cell leaders, and even commune councilors.11 29. These problems could probably find a solution in better distribution of responsibilities, particularly through more frequent delegation of powers. Excessive activity on the part of the burgomaster may often explain the passivity of his potential associates, either elected or technical, who are rarely asked to take on tasks requiring initiative or responsibility. A functional distribution of work and documents within the team of elected commune officials would certainly help to mobilize flagging energies. B. Functions of the Commune (a) Commune development 30. The "communal develc;nment" is a predominant theme in the national debate on the country's economic and social prospects: "If we are to avoid a situation in which excess rural manpower, mainly the younger generations, begins to pour into the urban centers -- a process of spontaneous urbanization is already starting in Kigali -- which could give rise to economic, social and ultimately political problems for the authorities, we must develop the instruments and find the resources needed to limit and control this process of migration to the towns by acting at the commune level" (MINIPLAN document). ll/As MININTER has noted, "There is urgent need to eliminate all passivity among these community leaders." - 11 - 31. In this context, the formula "the commune, the basic unit of development," is faithfully reproduced in every official document since the promulgation of the Second Republic A Communal Action Plan ("Plan d'Action Communal' -- PAC) was formulated in 1975 as a "sort of bible presenting a catalogue of the operations to be carried out to facilitate the development of the communes" (MININTER document).12 The objectives of PAC in terms of training and sensitization, guidance, formulation and project implementation were multiple and varied, and opened the way for the commune to carry out a vast range of activities without clearly specifying the respective financial responsibility of the communes and the central government. In 1984, the second Communal Program (prepared by MINIPLAN) recognized that the specific role of the commune "is nowhere embodied in the law" and that "action by the communes beyond their purely executive administrative function" remains in part "an objective, more than a reality." This document identified as major issuest (i) the lack of qualified personnel; (ii) no clear delineation of responsibilities between the central government and communal services; (iii) high budgetary burden on the communes due to salaries of staff which should have been paid by the central government; and (iv) inadequate involvement of the communes in centrally managed projects. 32. The functions performed by the communes may be grouped under three major headings: Mi) Political sens.tization ('encadrement"), as executing agency for the Central Government and as spokesman for the local community. In this regard, the role of the "cellulesa as the basic units of the commune merge with the organization of the Party; (ii) Administrative management (such as security, registry of civil status, local tax collection, etc.), by coordinating the services provided directly by the commune and by the various technical ministries; and (iii) Identification and implementation of economic and social development activities, both those pertaining to traditiona' commune responsibilities (social infrastructure, road maintenance, agricultural extension, etc) and those stemmiag from voluntary initiatives (direct implementation of development projects, collaboration with outside parties, e.g., "projects," NGOs, or support for local group and cooperatives). 33. The last of these functions -- the commune's economic and social activities -- is the topic of the present study. It is highly varied in nature; in the following sections only two of the commune's multisectorial functions -- execution of development projects and management of communal assets -- are reviewed. The second part of this report analyzes the actions of the communes within their institutional and sectorial framework. 12/A non-exhaustive listing of the actions to be undertaken as proposed by the PAC would include: recruitment of personnel, purchase of a commune vehicle, building of earth roads, electrification, water supply, infrastructure for commune health care and social services including a pharmacy, construction of a CCDFP (par&. 173), construction of a silo, 2 garage, a post office, an official guesthouse, and staff housing. - 12 - (b) Project Implementation 34. This is 8 priority objective for the burgomaster1 in part because the commune's resources -- both financial and technical -- are limited. By including their commune in a development project, or attracting an international development agency or an NGO to participate in their local projects, the burgomaster tries to overcome such constraints. This is particularly the case for small-scale, low-cost projects, with a clearly defined timetable and designed to produce tangible results. To this end, the burgomasters establish personal contacts (with the embassies, for example) in Kigali. These praccices, which are actively promoted by the supervisory authorities and appreciated by the development cooperation agencies because of their flexibility, are however a source of inequality among communes. In the final analysis, the amount of externally financed activities in the commune depends on the network of personal relationships built up by the burgomasters and members of the Technical Commission (para. 26) and the extent to which they can adapt to the conditionality of these interventions. 35. If the development projects involve an entire region or prefecture, the role of the burgomasters is more limited, and choices are generally made at the central government level. Even though the Government aims at providing each prefecture and commune with a project, the inequalities are more pronounced in the distribution of national projects.13 It is rather rare, in this case, for the communes to be considered as full partners in project participation or be actively involved in their preparation. 36. The dependency on "projects" (externally-financed andlor managed) is particularly evident when the commune takes over the operation and maintenance of a project after its completion.14 The degree of engagement by the commune is then proportional to its involvement in planning and implementation. Often the communes are obliged to assume responsibility for infrastructure facilities and other works that are far beyond their capacity and at times not even economically justified. For example, the labor force to be mobilized through Umuganda can be rapidly overwhelmed by new project-originated responsibilities such as maintenance of roads and forests. (For more comments on agricultural projects, see paras. 107-110). 37. The local population does not, in general, question the nature of the "projects" to be carried out, provided: (i) they participate as paid labor (thereby earning extra cash income); (ii) land developments do not aftect their farm holdings (requisition of land or encroachment for infrastructure or other 13/Over the 1982-84 period, almost 90 percent of the Government's total capital inveetments (not included in the communal budgets) went to the four prefectures of Kigali, Ruhengeri, Gisenyi and Cyangugu (which together contain one-quarter of the country's population). In contrast, the prefecture of Gitaramal the most populous prefecture after Kigali, received only 0.16 percent. 14/Frequent setbacks are aoted in project evaluations and in MINIPLAN documents. - 13 - works); (iii) the works can be reversed (erosion control measures), and (iv) most of all "the projects" do not involve compulsory participation in the form of labor or result in heavy financial charges (payment for water supply) (paras. 202-207 on Umuganda).15 (c) Management of commune assets 38. Every commune has some assets which can produce monetary income. The commune uses these resources at its own discretion or in collaboration (at times in an atmosphere of conflict) with other economic agents. Such is the case with small mineral deposits, for which it can grant mining rights. The communes are also in charge of two types of land reserve: (i) forests (paras. 141-142) and marshlands; the communes' responsibility for the development, management and exploitation of these lands is usually contested by the Ministry of Agriculture (MINAGRI) (para. 140); and (ii) vacant land whose appropriation or allocation is liable to provoke a sharp reaction among the local populace. In practice, it is usually the commune councilors who deal with land matters. There are also communal nurseries and demonstration fields, which the population often resent because of their establishment on scarce communal land. 39. Overall, the interest that the local population take in communal property and in its development -- to which they are required to contribute -- will depend primarily on their perception of the communal authorities and on the local social "climate". At times, the benefits are not evident; all too often the population seem to disregard or underestimate both the immediate and the longer-run advantages they can draw from the investments maL and work performed. 40. Another aspect of the communes' involvement in economic development is the pursuit of "self-standing' activities. Some operations (slaughterhouses, breeding stations, fruit-tree seedlings, farm inputs, etc.) have traditionally been part of the commune's activities, as they have usually been promoted by "projects". Likewise, the communes have undertaken some activities, such as site preparation, which may have favored specific individuals. Despite such cases, however, burgomasters are generally not very highly motivated in promoting the productive sector. A burgomaster would not be very interested in raising the productivity (thus the income) of individuals until this translates into higher tax paying capacity. But in the current fiscal system, communal tax revenues are not linked to individuals' outpuc or wealth. 41. The situation is entcirely different if the burgomaster and the commune are dire_tly involved in industrial or commercial operations. Then the temptation is great to take direct charge of certain activities, in competition or at the expense of individual or collective initiatives (grain and cassava mills, pharmacies, and more recently gasoline stations). However, the law is clear in this respect: "The management of industrial or commercial activities by the communes is subject to the prior authorization of the Prefect. Sach operations shall be organized in the form of autonomous Authorities ("regies") to be operated independently of the regular communal services." 15/nAs long as the projects are hiring people, families can pay their taxes," but 'Who will pay the salaries of the commune workers currently paid by the project?" etc. - 14 - 42. Significant reservations have been expressed at the highest levels in various Ministries regarding the predilection of certain burgomasters for purely commercial ventures. As a result, an earnest debate has been opened on the appropriate nature of and place for communal initiatives vis-a-vis other potential operators. The second draft Communal Program by MINIPLAN stressed that the "questionable eligibility and doubtful economics of direct management of economic and commercial assets by the communes should lead to immediate search for alternative solutions." Referring in particular to forestry operations (footnote 58), MINAGRI would like to see precise rules to govern the management of communal property, as certain burgomasters have no hesitation in resorting to clearcutting in order to finance a given project. MININTER, for its part, is alarmed by the lack of clear and strict accounting of the communes' commercial undertakings. The difficulties in applying the current laws, lack of knowledge of local practices (which prevents the Prefects from taking action) and, last but not least, the ambiguous attitude of the national authorities ("The communes must, after all, have some room to exercise their autonomy") mean that there is a need to solve this issue and precisely define the commune's authority in this area. 43. The direct management of commercial assets by the communes does not constitute an optimal use of their resources. One can only agree with the following official statements: "Most of the time, such a procedure (direct management) is less attractive than operating through concessionaires, where the commune -- after negotiating a specific economic project -- remains just one of the shareholders. Such solutions can be more productive in the long run as a source of revenue" (MINIPLAN); "The commune should restrict itself to promoting initiatives by individuals or companies' (MININTER). - 15 - CHAPTER III. THE FINANCIAL FRAMEWORK 44. The substantial financial resources mobilized by the commune have helped it acquire its current standing, but they still fall short of enabling the commune to perform its vast range of assignments. First, the communnsa, enjoy a rather remarkable degree of financial autonomy, and overall they ass-me a growing role in the public sector's financial flows. Second, local taxation is relatively flexible; its structure is highly varied, but needs some modifications. Third, not only are the communes increasing their current activities, but they are also contributing to the financing of development expenditures. Finally, the respective roles of the communes and of the Communal Development Fund (FDC) with regard to communal investments should be reviewed. 45. Local finances play a significant role in mobilizing domestic resources. In the 143 communes as a whole, the revenues (current and capital) mobilized by them on their territories may have amounted to about 1.3 billion RF in 1985, equivalent to 8 percent of central government tax revenue. They finance at least 80 percent of total budgeted expenditures (current and capital). The Central Government provides some subsidies (for personnel) and capital grants for specific projects. In addition, some communes receive grants (from NGOs) and loans (from FDC and banks) but external aid is not normally accounted for in communal (and more generally in government) finances. The mission has calculated the totals and the trends of revenues and expenditures (both current an.t capital) over a six- year period (1980-85). They are given in Table I. Further analysis of these aggregates has been on a sample of 18 communes (11 rural and 7 urban). It is presented in paras. 62-70, but cannot claim to give a comprehensive picture of overall communal finances. Table 1 Consolidated Budgetary Results for All Communes (RF million) 1980 1981 1982 1983 1984 1985 Current Revenues 697.8 736.8 834.8 941.5 1,032.5 1,067.7 Current Expenditures 734.5 857.7 912.7 987.2 1,041.2 1,044.0 Capital Revenues 224.0 276.5 335.3 366.5 465.3 574.1 Capital Expenditures 245.0 234.8 314.8 331.9 420.1 519.1 Overall Deficit/Surplus -57.7 -79.2 -57.4 -11.1 36.5 78.7 Me=o Item: Government subsidies & grants * 40.0 80.0 103.0 135.0 135.0 65.7 *Although these transfers to the communes are recorded in Central Government's current budget, they have been in part used to finance investments such as communal buildings (rf. Table II). Source: MININTER and mission calculations. - 16 - 46. One of the difficulties in analyzing communal finances is that special levies (para. 57) which have the characteristics of a tax, as well as school fees (para. 163), are shown as a capital receipt. Their amount varies considerably from commune to commune (para. 67). Current revenues proper mostly come from local taxation (paras. 49-56) while government subsidies account for a low proportion (para. 63). On a per capita basis and in constant francs, total current revenues tended to decline slightly over the six-year period, owing to an annual decrease of 0.8 percent in rural receipts (while urban revenues rose by 3.4 percent per annum). There are wide disparities among prefectures, and among communes within the same prefecture. 47. Recurrent expenditures have grown less rapidly than current revenues, and in 1985 the overall chronic deficit in the current budget turned into a surplus which was equal to RF 25 million (or 2 percent of total revenues). The urban communes show the highest rates of growth in both revenues and expenditures, with a budgetary surplus almost every year since 1981. On the other hand, the rural communes, had a chronic deficit until 1985 (which has been, however, declining steadily since 1981). This relatively favorable situation was in part due to the assumption by the Central Government of some communal debts. For the communes, one of the ways to adjust their expenditures to their revenues has consisted of lay-offs of communal employees. 48. Overall, communal investment expenditures increased from RF 245 million in 1980 to RF 519 million in 1985, equivalent to 50 percent of recurrent outlays and to roughly 17 percent of the Central Government's development budget. Per capita investment expenditures rose sharply during the period, at an annual rate of 14.6 percent. Their growth rate was about three times higher in the urban than in the rural communes.16 Moreover, as the communal budget does not include investments by NGOs and foreign aid agencies (as well as the work undertaken by Umuganda), the actual investment effort by the communes is much greater than the budget accounts would indicate. A. Commune Taxes 49. While the Central Government determines virtually all the outlays made by the communes, its financial contribution (subsidies and grants) is relatively small (7 percent of communest total revenues for the period 1980-85) and the communes are largely responsible for financing their expenditures. The Central Government has assigned certain taxes, known as "general taxes' ("impositions generales") to the communes, while leaving to the Communal Council the power to establish communal taxes, provided the tax base does not overlap that of the Central Government. This well-founded limitation has forced the communes to be creative in diversifying and broadening their sources of revenue. As a result, communal revenues are highly varied: in addition to the general taxes, the commune levies a series of local duties and special levies (or "umusanzu"). 16/Although this rate of increase slowed, sometimes sharply, in half of the prefectures, 80 percent of the prefectures showed a rise in the case of rural communes. This probably reflects the efforts of some rural communes to catch up in regard to community infrastructure (notably schools, health centers and administrative offices) and saturation of the absorptive capacity of certain urban communes. - 17 - (a) General taxes 50. There are two main general taxes, a head tax and a cattle tax, plus a tax on bicycles, and documentary fees ("droits de chancellerie"). For the 18 communes in the sample, these revenues represented 42.3 percent of total current revenues in 1980-85 (27.5 percent in the urban communes and 48.5 percent in the rural communes). Their relative importance declined steadily over the period. 51. The minimum personal tax ("contribution personnelle minimum' -- CPM) is an annual capitation tax of RF 400 payable by every male aged 18 and above and every woman working earning cash income outside the household. Its rate has not been changed in twenty years; as a result of inflation, its proceeds have declined substantially in real terms. Given the age distribution of the population and the exemptions (members of the military, police, clergy, students, etc.), only 20-23 percent of the population are actually liable. Moreover, the tax collection rate is rather low, articularly in the urban communes, though it has improved in recent years.37 Nonetheless, even if the amount collected were to rise 20-30 percent, the impact on the commanal budget would be marginal short of substantial increase in the rate. 52. The cattle tax ("taxe sur le gros betail") is payable at the rate of RF 250 per animal of 12 months or more. Although the national herd is believed to be declining in numbers (a finding that the recent Agricultural Survey seems to dispute), the outturn from the tax has remained relatively stable over the past five years, which may indicate somewhat of an improvement in compliance rates. 53. It is indicative of the Rwandese pragmatism that capitation (CPM) and livestock taxes are still being levied in Rwanda. These taxes which had a heavy colonial connotation (and were therefore often abolished in Africa at Independence) are being reintroduced in some African countries. Nonetheless, in a modern fiscal system such taxes should be integrated into tthe national income tax. This can be achieved only over the long-run as Rwanda's rural economy becomes more monetized. The CPM and the tax on livestock will remain for several years one of the main but progressively less important sources of revenue for the communes. In the interim, the rate of CPM could be revised to include special levies (para. 57) and be graduated according to the degree of wealth and monetarisation of communes and/or regions. (b) Local duties 54. The large variety of these taxes illustrates the autonomy and initiatives of the burgomaster and Commune Council; there were more than 30 such duties in the Commune of Ngoma in 1985. The- are mainly taxes on economic activity and can be categorized as followss (i) "occasional' taxes (on markets, for example); (ii) "periodic" taxes (on certain activities or services); and (iii) 'urban" taxes (often associated with services provided to private, commercial and industrial plots of land). 17/In the sample, the collection rate (for the period 1980-85) was about 69 percent for the rural communes and 35 percent for the urban communes. However, in 70 percent of the communes CPM revenues (in nominal terms) increase faster than population growth. - 18 - 55. Although the Communal Law prohibits the communes from taxing items already taxed by the Central Government, it is not uncommon that certain communal taxes may be at odds with national legislation. In fact, certain local taxes have much in common with the "patentee (business license tax) in that they all relate to the same economic activities. The "patentel is a national minimum tax levied annually on industrial or commercial enterprises. When the enterprise is individually owned, its amount is determined on a lump-sum basis, taking into account the location of the means of production (and the type of construction). It is recommended to eliminate the ambiguities between these local taxes and the "patente", as taxpayers often fail to understand the reasons behind their assessment and contest the amount payable. 56. Local taxes are the category that has risen most sharply in recent years.18 In the future, they are likely to increase in importance and enable the communes to finance a substantial part of their development. The yield from such taxes is directly linked to the wealth of the commune, as this is entirely appropriate. However, a word of caution is needed here since the communes seem to be meeting their financial needs largely through these taxes without measuring their longer-term impact. Recourse to this form of taxation should not become excessive, as it may result UOL only in fraud but also in a decline in economic activity and thus ultimately in reduced contributive capacity. This may, for instance, lead farmers -- who are relatively poor -- and vendors, to avoid markets where tax rates are prohibitive. (c) Special levies 57. In order to finance their capital investments,19 most of the commures have -- not totally in line with current legislation -- introduced a system of special levies ("umusanzu" which are qualified as 'voluntary-compulsory") for specific purposes, such as the construction of schools, a communal office, a health center, etc. Sizable sums have been collected in this manner (para. 67). These levies are, however, widely criticized by the population, as well as the Central Government, who fear that the funds collected are improperly budgetized and/or being spent for other purposes. (d) Other revenues 58. Other sources of commune revenue are: (i) income from communal assets (rentals, sale of seeds or forest products); (ii) income from commercial activities (flour mill, service station, etc.); and (iii) incidental and miscellaneous revenues (fines, payments for use of the communal truck, medical charges, etc.). There is a disquieting trend for some communes to resort to higher fines in order to improve their finances. While this method can provide a temporary solution to a commune's financial problems, it carries with it the risk of alienating the population over the long run and prompting taxpayers to 18/In the sample, the annual rate of increase was 12.5 percent for the rural communes and 22.8 percent for the urban communes. 19/The proceeds of these levies are recorded in the investment budget ('budget extraotdinaire"). - 19 - resist paying for commutal services and contributing to development projects.20 (e) Toward a land tax 59. To diversify the communes' sources of revenue, the Government should start the process necessary for the introduction of a land tax. Such a tax is local in nature; it bears essentially on the additional value accruing to the owner of a piece of property from the provision of communal infrastructure. It is intended to fiuance the maintenance cost of that infrastructure, and is preferable in this regard to a multiplicity of taxes earmarked for specific services. Currently, the Central Government taxes the surface of buildings, and the surface of unimproved land. These two taxes produce little in the way of income (R* 125 million budgeted for 1986, or 2.5 percent of the total revenues of the "Direction Generale des Impots") and take no account of the quality, the use, and the location of the property. 60. These taxes should be based on a given property's potential commercial value. As far as the property tax is concerned, the marked differentiation of the urban space (between for example, high and low income residential areas) should make it easier to prepare value estimates based on the type of neighborhood. To ensure greater social justice, the property tax should take account of the following factorss (i) the intended use of the building (residential, rental, industrial or commercial); (ii) its location within the neighborhood; (iii) the type of construction (permanent, semipermanent, nonpermanent); (iv) amenities (water, electricity); and (v) surface. As a declarative tax to be occasionally verified by the authorities this would not require full cadastral survey, but merely a list of the taxable properties. It has been roughly estimated21 that if such a tax were applied in Kigali it would yield about RF 300 million, i.e. more than double the amount budgeted for the entire country. 61. It is, therefore, recommended that the Central Government transfer to the communes, either in whole or in part, the two taxes currently levied on property. The communes are often better equipped to determine and tap the contributive basis and more motivated than the central authorities to collect a tax that represents only a very small proportion of government revenues. The recommendation is, therefore: (i) to review the rate structure, after first studying its incidence on urban households' incomes; and (ii) to establish a simplified cadaster along the lines of an urban property register. This tax system should be introduced in stages. Consideration could be given, in connection with the IDA-financed urban project in Kigali, to conducting a pilot project that could later be extended to other urban centers, and then to the rural communes. In the case of the latter, a specific study on the impact of a land tax would also be needed, as the economy in certain regions is not much monetized. 20/The system of fines has become so abusive in some communes "that the people have been asked to listen to radio broadcasts put out by MININTER to inform them of their rights." 'The commune should not seek to solve its financial problems at the expense of the population." Statements from the seminar for Subprefects, Kigali, November 24, 1986. 21/Based on the 1982 housing survey in Kigali, and assuming 80 percent collection -rate. - 20 - B. Communal Budgets (a) Current budget 62. Most administrative expenditures of the commune are financed out of the commune's current budget. The Organic Law initially assigned an impressive list of expenditures to the communes; since then, with additional tasks, these outlays have become even more diversified.22 In the sample of 18 communes, outlays on administration and police accounted for almost half of the current budget; expenditures on maintenance of communal assets (public works), on agriculture (including stockraising, water resources and forestry), and on education each accounted for more than 10 percent. Expenditures on agriculture, education and health care are discussed in detail in Chapters V and VI of this report. It should be noted that debt service is included in the current budget: its share varies from 1 percent to 14 percent depending on the commune, with urban communes carrying the heavier debt burden. 63. Current subsidies from the Central Government are usually accorded to pay salaries of communal workers. These subsidies are relatively small; in the 18-commune sample, they represented only 4 percent of the revenues received by the communes in the period 1980-85. Their incidence on the ordinary budget is declining as total government contributions have been reduced in the last two years and are increasingly in the form of capital grants (over 75 percent). 64. It can be estimated (from the sample) that the incidence of local taxation23 on the average household in 1985 amounted to RF 1,576 in the rural communes and RF 3,752 in the urban communes. On the basis of awverage monetary incomes as estimated from household budget surveys, the communal fiscal burden24 both on rural and urban households would be around 5 percent. (b) Investment budget 65. The obligations of the communes with respect to development expenditures are not specified in any legislation. Strictly speaking, even a Presidential Order (Arrete Presidentiel) cannot require the communes to incur expenditures; the Organic Law states that a law is needed for this purpose. Nevertheless, a whole range of investment outlays have de facto become obligatory for the communes. Primary schools are an example. The Law regards construction of such schools as essentially the responsibility of the Central 22/The communal budgets contain a number of administrative expenses that have been decided by the Central Government but are not sanctioned by law. Svch is the case, for instance, with the salaries of the youth leaders ('encadreurs de la jeunesse') who have been recruited by the central authorities and placed on the communes' budgets without proper legal authorization (stated in MINIPLAN documents). 23/Includes current revenues, less government subsidies, plus special levies in the investment budget 24/It should be noted here that the total financial contribution made by a family includes -- above and beyond local and central government taxes -- payments to the MRND, school (minerval) and other fees. - 21 - Government which can only be delegated to communes on an individual basis. The General Regulations for the implementation of the Law indicate, on the contrary, that school construction is to be undertaken primarily by the commune. Following the public discussion of educational reform, it has then become customary to consider the construction of primary schools as a mandatory task for the commune. The expansion of primary education requires a rapid pace of construction, about 700 classrooms each year on a national scale. Based on our sample, the communes have actually exceeded this goal by building roughly eight new classrooms a year. 66. The communes' budgets comprise a long list of investments: new construction and major repairs of school centers (primary school and school workshops), centers of literacy and non-formal training (CCDFPs), communal offices, sector offices, integrated rural and artisanal instruction centers (CERAIs), commune roads and bridges, plus housing for government officials and communal employees. In addition, the commune is required to set up a cooperative silo, a warehouse, a slaughterhouse and an animal breeding center, and purchase the vehicles required for their operation. Urban and rural communes have spent a relatively similar share of their extraordinary budget on administrative facilities and school buildings respectively (38 and 41 percent for the urban communes and 29 and 31 percent for the rural communes). However, urban communes allocated comparatively less to the construction of health care center (1 percent as compared to 7 percent spent by the rural communes) but more to the purchase of goods, vehicles and furniture (12 percent compared with 5 percent). 67. Self-financing plays a significant role in the investment budget, as "extraordinary' revenues include locally raised resources such as proceeds from sales of communal property and, in particular, special levies (umusanzu) and school fees. The "umusanzu" are paid by the residents of the commune to help meet specific "extraordinary' expenditures (construction of a school, an office, a market, a health center, etc., and at times in lieu of Umuganda). In the 18- commune sample, 30 percent of investments in 1980-85 was financed out of these special levies. There are, however, substantial differences among communes: some did not resort to this form of taxation at all, while others have funded more than 70 percent of their investment budgets from such levies. 68. Such disparities are not disquieting as such; however, the scale and multiplicity of the construction tasks assined to the communes, and the extent of the burden on taxpayers strongly suggesi the need for the communes to program their investment expenditures. Such programming would, at the same time, make it possible to formulate sectoral priorities in infrastructure* to spread out personal contributions, and to rationalize the search for alternative (external) sources of financing. 69. The external resources available to the communes consist essentially of contributions from the Central Government or other public organizations, and loans. Contributions from the Central Government are granted in connection with specific investment outlays (mainly for cons.truction). These grants whose alleged objective is to foster equity between rich and poor communes oftvn transit through the Communal Development Fund. Some commune budgets also benefit directly from the assistance of NGOs or of other bilateral aid organizations. - 22 - 70. Borrowing by rural communes is virtually nonexistent, whereas urban communes, which enjoy a stronger revenue position, have borrowed on a regular basis. There are at present two financial institutions in a position to make loans to communes, namely the "Banques Populaires" (People's banks) and the Communal Development Fund. When the latter provides its guarantee, the 'Banque Rwandaise de Developpement" can also extend loans. C. Communal Development Fund 71. The Communal Development Fund ("Fonds de Developpement Communal" -- FDC) was set up in 1977 with two appropriate goals: (i) to effect equalization among communes in order to reduce territorial disparities, and (ii) to give communes access to credit, either in the form of direct loans or by establishing a Guarantee Fund to enable communes to borrow from banks. Rules were laid down regarding the allocation of resources between grants, loans and guarantees according to their origin; but chese rules have been observed only in part (para. 74). The Decree-Law establishing FDC endowed it witk. legal personality and financial autonomy; but FDC currently operates merely as a department of the Central Government (MININTER). (a) FDC resources 72. The resources of the FDC consist of: (i) annual contributions from the communes, set at 10 percent of their current revenues;25 (ii) subsidies and grants from the Central Government, and aid agencies; (iii) interest and repayments on loans and investments; and (iv) if applicable, deposits of "special funds"26 by the communes. A reconstitution of financial flows from the Fund's inception through the end of 1985 shows that 39 percent of its resources were derived from commune contributions, 51 percent from government funds, and 9 percent from such external donors as Swiss Aid, USAID and FAC. 73. The Decree-Law establishing FDC stipulated that the contributions paid in by the communes (constituting FDC's 'own funds") were to be employed as follows: (i) 50 percent for loans to the communes; (ii) 20 percent for grants under the Commune Equalization Fund (Caisse de Perequation Communale); and (iii) 30 percent for guaranteeing the loans contracted by the communes from financial institutions -- with the proviso that total guarantees could not exceed four times the Guarantee Fund's assets. Funds received from the Central Government are usually earmarked for specific projects. Lastly, the Decree-Law provided that FDC loans could only be used to finance productive projects. 25/As these contributiius do not pass in full through the communal budget, it is difficult to track them at the communal level and hence to determine each commune's net accounting balance with FDC. For instance, the CPM payable by government officials and commune employees is withheld at source and transferred directly to FDC; also, license fees on drinking establishments are paid to the prefecture. These practices probaly reflect distrust on the part of the Central Government vis-a-vis the communes. 26/These represent various cash funds that are somewhat in transit. - 2:l - (b) FDC operations 74. The Fund had difficulties in distributing its operations among loans, grants and guar3ntees as prescribed by the Decree-Law. Of FDC's operations financed out of its own funds, 50 percent went to grants (compared with a target of 20 percent), 46 percent to loans (compared with 50 percent), and 4 percent to the Guarantee Fund (compared with 30 percent). The high proportion (70 percent) of investments in administrative facilities (largely communal offices) is attributable to national campaigns ordered by the Central Government (a communal office in each commane by 1986, for example); but it also reflects the difficulty the communes encounter in submitting well prepared development projects to the FDC. Table II Distribution of FDC Grants and Loans by Type of Project*/ 11977-85) Type of Project Amount Percentage (RF million) Administrative Facilities (commune offices and vehicles) 411.3 70 Water supply, Electrification 27.3 5 Health Care 34.3 6 Education 14.4 2 Infrastructure 11.3 2 Markets 7.5 1 Artisanal Activities and Farm Cooperatives 14.4 2 Trucks, Machinery (tractors) 72.0 12 Total 592.5 100 */ Includes only projects financed out of both FDC's own funds and aid from Swiss Aid and USAID. It was not possible to break down Central Government grants, as no information was available. However, it is known that they were used mainly to settle debts owed by the communes and to finance communal offices. 75. The FDC's administrative capabilities seem to be limited. First, the appraisal and follow-up of project documents is not only protracted but also inadequate. Second, the accounting records do not adequately reflect the Fund's current financial situation because debt defaults have not been properly documented. Some accounting practices27 have erroneously led to a questioning of the double-entry bookkeeping system but it would be a mistake to give up that system as it is the only way to guarantee the regularity of the adcounts and adequately reflect the Fund's operations. Third, the lack of forward planning makes the FDC's financial situation unclear, if not precarious. There is not a single annual report that indicates the amount of the FDC's commitments as guarantees for communal loans. According to mission's estimates, this would 27/Grants are shown as liabilities of the FDC, whereas they are in fact uses of funds, and certainly not losses. - 24 - amount to approximately RF 130 million as of end 1985. On that date, only 17 percent of the required amount for the Guarantee Fund had been deposited with the Savings Bank ("Caisse d'Epargne"). This situation is likely to cause serious financial problems in the future because the percentage of borrowers availing themselves of FDC guarantees is particularly high and recovery rates on direct loans are relatively low. (c) The FDC and Development Finance 76. A far-reaching reform of FDC is needed to better organize the intermediation of centrally provided resources.28 To enable FDC to achieve its goals and to play fully its role as a financial Intermediary, it must first recover the legal status and financial autonomy conferred by the Decree-Law. Without eliminating Central Goverrment supervision, a better balance could be struck between Central Government representatives (ex officio directors) and representatives of local authorities (elected members) on its governing bodies. 77. It is also necessary to strengthen FDC's management structure not only to supervise its current operations more adequately but also to make the necessary adjustments for its future activities (Rural Water Fund, for example). There is need for staff training, provision of materials (such as microcomputers), as well as a review of lending and accounting procedures. 78. It is also recommended to (i) review the procedures for allocating FDC's own resources between grants, loans and guarantee funds to take better account not only of the poverty and needs of each commune but also of its relative efficiency (as indicated, for example, by the recovery rate of local taxes) and its creditworthiness (including timely debt servicing); when new allocation criteria have been established, they Till have to be respected; (ii) the Guarantee Fund will need to be properly provisioned, and interest received could meet at least part of FDC's operating expenses; (iii) in no case should the assistance provided be concentrated on a small number of communes but should take account of objective indicators of each commune's relative poverty;29 and (iv) the FDC should finance more relevant projects of economic and social infrastructure as proposed by the communes themselves and not in majority those mandated as national goals. 79. The technical capabilities of the communes also need to be Improved, as they have somewhat inhibited FDC's operations. Technical responsibility should remain with the relevant ministries, but these should consider ways and means to improve their assistance to local authorities. Consideration could also be given to setting up a project study and appraisal group within the FDC, which might be financed by an NGO study fund. Such steps could modify current practices and improve the image FDC among the communal authorities by transforming it into a flexible and dynamic institution ready to assist the communes in formulating and financing their own ptojects. 28/The recent decision to open a "window" for the Rural Water Fund (Fonds d'Hydraulique Rurale) within the FDC shows the importance that donors and the Government attach to the FDC. 29/The statistics available at the commune level regarding population and revenues are remarkably detailed even though they are not systematically tabulated in any central document. - 25 - CHAPTER IV. THE SOCIOECONOMIC FRAMEWORK 80. The commune is currently in the middle of a transition. It has been in existence for more than 25 years and still is, in many respects, an administrative unit of the Central Government; at the same time, however, it is evolving as a decentralized community with growing responsibilities and autonomy. It is becoming a socioeconomic reality and a frame of reference not only for public sector activities but also for those of other economic agents. From a purely institutional (i.e., administrative and political) viewpoint, the commune has considerable means of action at its disposal. In addition, as we have also seen, the financial resources that it mobilizes play an essential role in establishing its relevance. In the final analysis, however, it can serve as an active agent of development only to the extent that it is recognized as such by its various partners. This would require a redistribution of the responsibilities between the Central Government and the communes, limitation of communal powers vis-a-vis the population, and a better adaptation of public interventions to local conditions. The definition of an appropriate strategy in these regards will reinforce the existence of the commune as a socio-economic reality. This last point will be the focus of the present Chapter and lead to some general conclusions. 81. The initial development strategy of Rwanda can probably be qualified as "direct intervention by the State alone" (a strategy inherited from colonial times). This was later replaced by a regional "project' approach based on iategrated development strategy. Now, we are witnessing a strengthening of a "communal' approach, resulting in a greater diversity of actions brought closer to the local populations. For such an approach to be successful, one should reflect on the role of the commune, with a view to limiting its activities to those that should be within the purview of public auth4orities. The existence and vitality, both present and potential, of the pri'aTte sector, of local groups (both occupational and other) and of users' associations should serve as the basis for this relatively new development approach. A. The Commune in Relationship to Other Parties (a) Relationship with the Central Government 82. The burgomaster is primarily the last link in the Central Government's line of command, the executor ot policies and decisions formulated on higher echelons. It is his dutv *, implement the decrees, circulars, decisions and instructions issued nationwide. Orders emanate from both the central and the regional level: (i) The Ministry of the Interior and the Party play a decisive role in the management of the communes, and the burgomasters take their cue from the demands and requirements of these authorities; and (ii) at the prefectural (and, increasingly, subprefectural) level, the application of national directives (particularly those of an explicitly political nature) is closely and regularly monitored, and the burgomasters are frequently summoned to attend seminars and other meetings. - 26 - 83. All the burgomasters are in agreement that their assignments are heavy, burdensome and disproportionate to what the Central Government provides in the way of staff, facilities and logistic support. They feel that their initiatives are deployed mainly to find resources locally to meet their obligations vis-a- via their hierarchical superiors.30 Lack of adequate financial resources are aggravated by the absence of formal criteria and a standard framework for the allocation of government funds to the communes. This results in disparities among communes and a certain bitterness among burgomasters: "Central funds are bestowed as gifts or rewards, even when they are not merited." This sense of rancor is heightened bv the fact that a burgomaster's room for maneuver vis-a- vis the prefectural and national authorities, and his ability to "interpret" circulars and directives from the center, is often directly proportional to the "pull" and influence he has in the capital. (b) Relationship with communal employees 84. The institutional framework has an important impact on the management of the commune's technical personnel. Some are paid out of the commune's budget; but all are rep,rted on by the burgomaster rather than by their technical ministries (to which they are attached). Their iwork is coordinated and supervised by the commune.31 For most of these employees, their future is determined by the commune's tax revenues. Irregular salary payments, sometimes running several months in arrears, underscore the precariousness of their Jobs, particularly in the poorest communes. Inevitably, therefore, many of them find themselves closely involved in simple policing tasks and implementation of regulations. There is also an obvious inequality in treatment among the various categories of communal personnel. Municipal police officers clearly occupy a privileged position and are well looked after by MININTER. The recent wave of layoffs (which affected a sizable number of agricultural extension agents)32 clearly demonstrate what the communal administration's priorities are. (c) Relationship with the population 85. The population perceives the burgomaster first and foremost as the representative of the Central Government, whose authority and responsibility extend to every aspect of rural life. But the burgomaster's involvement is primarily in the areas of resource mobilization and controls: collection of taxes, and special levies, control of land use and cattle, compulsory labor work, etc. Supervisory activities ("interventions d'encadrement") in 30/"Being a burgomaster means having a maximum of duties and a minimum of resources"; "The burgomaster is the only man who cannot pass the responsibility for implementing national directives on to someone else"; "We have to apply the circulars, but we are the ones who have the least freedom for interpreting them"; etc. 31/"For the burgomaster to have authority over them, most of them are not paid by the Central Government" (MININTER). 32/"It's their work that's seen as least useful" (a burgomaster). - 27 - agriculture are implemented through national campaigns (resulting in fines if not properly applied): maintenance of coffee plantations, erosion control œeasures, installation of compost pits, thinning of banana plantations, etc. (paras. 121-125). 86. The multiplicity of special levies (para. 57), the inevitable issues associated with their collection and, sometimes, the lack of information about their use33 may explain the reservations local populations have expressed on occasion when burgomasters have sought to enlarge the communes' area of activity. Any strengthening of the power of the burgomasters, any extension of the commune's supervisory, management and implementation capabilities, is generally seen by the population as heralding new demands in terms of money and/or labor. This attitude does not reflect systematic hostility nor refusal to cooperate, but is simply the result of past experience with development projects whose real benefits were not always apparent to the peasants.34 87. While the burgomaster has virtually absolute power, there are definite limits to its actions. Resistance to burgomasters' interventions may take various forms: (i) the main recourse is to appeal to higher authority (the Prefecture), a procedure used when a solid case has been built up and enjoys local support; (ii) mobilization of "traditional" structures (family, local groupings), with their internal solidarity that defies outside control; and (iii) passive resistance to communal staff's plan for "mobilization" in various areas. Moreover, in addition to individuals and groups of individuals, a major local counterweight to the burgomaster's power is represented by the other "institutional" powers, primarily the churches (which have the means to act in most areas of social life), but also NGOs and "projects," which can negotiate directly with the technical and administrative supervisory authorities. 88. The ways in which the commune's power is exercised and how it is perceived vary greatly, depending on the local setting. The poorer and more remote a commune, the more the burgomaster is personally identified with the commune while at the same time he feels virtually powerless in the face of populations accustomed to relying solely on themselves to resolve the most pressing problems. In such cases, all that a burgomaster can do is negotiate with the local notables on how to carry out the tasks and requirements of his office. 89. The situation is totally different when the commune is involved in wider trading and other relationships resulting in increased population mobility (trading of farm produce, additional income from artisanal and other activities, 33/"The various levies known as umusanzu burden the people because of the disorganized way in which they are levied"; "Certain burgomasters demanded umusanzu for a project that was unknown to the population, and collected it over a number of years; what was particularly disheartening was that after this period there was still no project to be seen" (Monitoring Report, MININTER). 34/"We must put ourselves in the people's shoes, otherwise we could be headed for disaster"; "We don't have time to wait for the farmers to come round to seeing what needs to be done." - 28 - health care, lengthy schooling, etc.). In suc a case, the communal authorities find it necessary to adopt more complex strategies vis-a-vis diversified and often better-off population groups (traders, government officials, artisans, truckers, etc.). (d) Relationship with local groups 90. Despite the apparent isolation and fragmentation of settlement patterns in Rwanda, social life in the rural areas is intense and numerous forms of association give concrete shape to mutual solidarity and community actions. The widespread presence of cooperative, associative and risk-sharing groups, which is considered to be one of the distinguishing features of the Rwandese countryside, is largely responsible for the vitality of local communities. There are well-structured farmers' associations in the form of cooperatives in the agriculture sector (paras. 111-112). There are also many farmeta' groups (for crop production, housing improvement, artisanal activities, etc.) about which little is 'nown. 91. A recent study by IWACU estimated the number of such groups at more than 30,000. Particularly widespread and important in some regions, these groups are subject to very little government control. This relatively new phenomenon seems to reflect the appearance of new community welfare networks as well as production interests. Because these groupings are very informal, it is difficult to assess how likely they might eventually become "institutional" partners in the development process.35 92. A number of studies were undertaken on these mutual help groups, and one conducted in association with the Kibuye agricultural project is of particular interest.36 It demonstrates how wide the variety of farmer organizations is. The oldest of them have existed for 15 years, although the majority were set up quite recently. They are small, generally with an average of 11 to 15 members. With respect to farming pctivitiP7s, foe groups are mainly concerned with plowing, preparing and sowing the land and only rarely with crop mainte,nance and harvesting. This being so, most groups come together only during the period of hard agricultural work. 93. Along with these farmers' groups there are the traditional "tontine" type associations which play an important role in mobilizing financial savings in rural areas.37 These associations are based on the principle of putting all the members' contributions into one "pot" and sharing the proceeds in turn. They may take any of the following forms: (i) working together on each member's field in sequence; (ii) rooftile tontines, with each member contributing a given number of tiles and receiving all the tiles when his turn comes up; and (iii) 35/The experience of the People's Banks with bridging loans ("credit de soudure") guaranteed by such groups would appear to provide evidence of this potential. 361M. Vianney and T. Klias, Les Organisations Spontanees en Milieu Paysan (Une etude menee dans la zone du projet agricole de Kibuye), IWACU, 1985. 37/Some of these associations have a highly developed structure, with a president, vice president, secretary, treasurer, and a policeman. - 29 - savings and loan associations, with each member benefitting in turn from the members' contributions.38 94. The more structured groups have not, however, been able to grow to any size except where the staff of outside agencies have taken on position of responsibility within the groups. While the farmers are thus represented "by delegation," their spokesmen often assume the triple role of initiator, organizer and guarantor. In most cases, the actual producers do not occupy the responsible positions that ought to be theirs, these posts generally being held by government officials, technicians, members of the clergy or members of foreign aid agencies. Moreover, communal leaders, like their hierarchical superiors, are by tradition eager to control any initiative that could become too independent. It is important to consider these constraints when seeking to ensure that projects or other interventions take root and are later able to pass on their functions (in total or in part) to these farmers' groups (sustainability issue). But it is even more important for the public authorities to modify their approach to private initiatives. 95. The communal administration has an important role to play in the emergence of new institutions. This is, for example, the case with the "Banques Populaires" (People's Banks), which, have been set up at the request of the local authorities. Their establishment is often contingent upon material support from the commune (the People's Bank will not open a branch in a commune unless the commune provides the premises). The involvement of the communal administration in the establishment of People's Banks -- while having the advanitage of facilitating its operations -- has frequently had the effect of giving the bank the image of a communal institution. Effective power on the bank's board is often exercised by the burgomaster (even if he is not the chairman) and by the communal councilors, particularly if the bank is located in the commune office. The current trend is to make the banks autonomous vis-a-vis the administrative authorities and to improve the representation of major depositors on the banks' boards.39 B. The Commune as an Economic and Social Reality 96. Although each commune has its own distinguishing features and is strongly marked by the personality of its burgomaster, all the observations made in the field show that the commune constitutes an extremely homogeneous political and administrative organization. Despite the limitations on the power 38/A recent study of five communes by IWACU revealed more than 1,035 such groups with an average of 12 members each. A study of a commune in Byumba shows that there are 95 tontines with a total membership of 5,300 and assets equivalent to RF 18.5 million (about US$ 210,000). 39/In all the bank boards, the representation of farmers has grown while that of commune councilors and commune officials has fallen. As to burgomasters, 22 were board chairmen in 1977 as against only one in 1983 (Ministere des Relations Exterieures, Republique Fransaise, Methodes de Mobilisatlon de l'Eparane Rurale dans les Pays Africains, December 1983). - 30 - of communes within a centralized government system, their importance in the country's development strategy and the recognized role of the burgomasters as required interlocutors for most development projects instill a real content into the communal organization from the viewpoint of external decision-makers. 97. The commune represents not only a highly active local administrative authority, but also (increasingly) an economic and social unit. The dense network of rural roads means that most "secteurs' are within easy reach of the burgomaster's authority, at least during the dry season. Access to schools and health care centers encourages the population to come to these focal points of local life. The recent construction of communal and sectoral offices is reinforcing this trend. The same is true of the construction of housing, markets, People's Banks and shops. 98. The commune is not yet, however, the most natural place for the population to identify with, the place where most economic and social relationships that determine daily life take place. First of all, the 'hilltops' (Ncollines de residence") are where ties of solidarity (both family and traditional) are most vigorous and waere links to the soil (rights to plots of land, to crops, etc.) are closest. The second "natural' echelon is where trading occurs. The market place may be in the "secteurs" or in the commune, but also outside the communal boundaries. The spatial, social and economic horizon for the great majority of the rural population is circumscribed by these two levels of reference. Household expenditure surveys40 show that virtually the entire rural population derives roughly three quarters of its income (mostly from bananas) from local trade or trade within the region. Only income from coffee and a fraction of income from wages (particularly the wages paid by 'projects") depend on outside sources. This can be characterized as a pattern of both extensive subsistence farming and local self-sufficiency. 99. Despite the variety of situations, communal officials (in particular the burgomasters) have been able, to some extent, to strenghten the commune as an economic and social reality41 However, this can only be better achieved if communal activities are more adapted to local needs. The "commune's power" can be embodied and exercised only within the dual confines of the "colline" (as a social setting) and the market place (as an economic setting), As an administrative, economic and social unit, the rural commune still is in many respects "remote" from the population.42 The burgomaster's freedom of action is essentially circumscribed by two factors: (i) the exceptional independence of 40/A. Guichaoua, 'Budgets et strategies monetaires des paysans au Burundi et au Rwanda," in L'Afrique des Grands Lacs, Revue Tiers-Monde, April-June 1986. 41/Nonetheless, not all residents are fully aware of what the commune's institutions are: a study in the commune of Mugusa showed that 46 percent of residents do not know what the Commune Council is -- even though it is the major decision-making body. 42/Relationships are indeed stronger within the "secteurs" making up the commune: "The secteur is the basic unit for encadrement (government campaigns), the basic collective work unit for Umuganda, and ultimately the social unit" (statement by a burgomaster). - 31 - the peasant farmers, 43 and (ii) the absence of resources provided by the Central Government. The commune is a point of reference only when and where it becomes a concrete reality. When the boundaries of certain "secteurs" ignore socio-economic logic and traditional trading relationships, and nearby urban centers exert a decisive pull, such a reality will be slow to develop. 100. In rural areas, the commune, with its 'secteurs' and hcellules"l, is certainly the level of government closest to the farmers. Rwanda's judicial system also empowers the commune to act as interlocutor and permits it to carry out all, or certainly most, types of action. Lastly, from the technical angle, while skill levels need to be strengthened, the existing structure is on the whole relatively satisfactory both as regards management and "encadrement". 101. In urban communes, the communal framework is viewed very differently and cannot be reduced simply to its administrative boundaries. The perceived spatial limits of the urban commune as an economic and social reality (particularly in the case of Kigali and the major prefectural capitals) fluctuate and change constantly as a result of urban migration and the rapid emergence of new neighborhoods. Similarly, peripheral communes are under the influence of urban centers and have little or no hold on their residents. One example is provided by the sugar mill at Rutongo (a commune peripheral to Kigali) where workers are mostly farmers from distant communes, while the residents of the commune go to work to Kigali. 102. In conclusion to this Part I of the report, it appears that within the communal setting, the country has achieved the mobilization of quite considerable human and financial resources, both externally and locally, and their utilization with some success to implement national priorities. The communes have been most successful in establishing a relatively efficient organization and in carrying out their administrative duties. The substantial amount of local investments has increasingly given a socioeconomic existence to the communes. Communal taxation, however, has .'sually aimed at revenue maximization rather than economic efficiency or equity. Together with widespread recourse to user fees, this taxation has provided the communes with a notable measure of fiscal autonomy. 103. In order for the communes to provide an even more appropriate setting for various development actions, some changes will be needed in the overall approach of both central and local authorities. First, there is a need to reduce the direct intervention of the Central Government and to decentralize its power more effectively. It is important to foster the implementation of sectoral policies through incentives to decentralized initiatives rather than 43/"Th" farmers are masters on their own land; you cannot force them to do anything; they do only what they want to do, when they want to do it, and they can manage totally without outside assistance." This statement sums up the frustration felt by most technical staff dispatched to "work in the countryside." Their difficulties are compounded by the fact that they fail to reach the female farmers, even though it is the women who do most of the work on the land. But this resistance to outside intervention can also be considered as a strength of the Rwandese farmers against the vagaries of climate and unreliable supplies of inputs. - 32 - the enforcement of centralized programs. Second, communal authorities should take greater account of the population at large, local groups, and other private initiatives, all of whom are very active in Rwanda. In certain areas (commercial activities for instance), the communes should reduce their direct involvement and devote more attention to encouraging, supporting and channelling these initiatives. Third, there is a need to better tailor public sector intervention to local conditions. Government campaigns, agricultural extension messages, and the community work result too often in specific instructions designed at the national level and strictly applied locally without enough attention being paid to the ecological, agricultural and economic conditions of individual communes. Numerous examples in support of these general points are given in Part II of the Report. i4, - 33 - PART II THE COMMUNE AS CENTER OF ACTIVITIES CHAPTER V. AGRICULTURAL ACTIVITIES 104. Rwanda is an extremely rural country, and agriculture is the very lifeblood of almost all its communes. Even the activities of the so-called urban communes consist primarily of the exchange of products and services from or to farmers. The rural communes increasingly serve as the frame within which farmers come into contact with the other agents of development and where programs are implemented to target them. The striking variety of actors and operations at the commune level is indicative of the importance and vigor of the commune as an institution. The commune must be continuously adapting itself to the farmers' demands, to the need for a rapid increase in agricultural production, and, in a more general sense, to the requirement of the economic and social development so dependent on that increase. If this necessary adaptation is to take place to its fullest extent, the commune has to evolve more rapidly from its initial concept of an administrative unit to that of a functional institution whose role should be delineated more clearly. A. The Actors (a) Communal personnel 105. For the provision of services to farmers, each commune has an agronomist and a veterinarian, who supervise the extension agents assigned to "secteurs" and ensure their liaison with prefectural staff.44 In some communes, there is also a forester. The clmmune agronomist and veterinarian45 are usually civil servants on the Central Government payroll. Each "secteur" usually has an extension agent (monagri) and a veterinary assistant (AIV), both paid by the commune. There are also a number of specialists attached to each commune to provide extension services in coffee-growing, bee-keeping, fish-farming and forestry activities. They work in conjunction with other communal personnel (female social and nutrition instructors, youth organizers, cooperative organizers) staffing the CCDFPs (para. 173). 106. In practice, the number of staff varies considerably from one commune to another, depending on: (i) the existence of "projects" (which finance all or some of its technical personnel); (ii) financial resources of the commune; and (iii) the secondment or financing of specialist 44|Each prefecture employs an agronomist and a veterinarian, whose authority extends throughout the prefecture. 45/There is, a shortage of fully qualified veterinarians, so that the veterinarian in many communes has received no more than on-the-job training. - 34 - personnel by other public institutions (parastatals, for example). As regards the impact of projects, for instance, the mission estimated that in 1984, communes with no project allocated approximately 12 percent of their current budget to agriculture, while those with a project allocated 6 percent only. In spite of the greater effort that the communes 'without project" demonstrate, they cannot finance all the personnel that they would need. The following table illustrates the theoretical and actual numbers of technical personnel found in the "typical" ten-sector commune that has no project and could only have one third of "expected" staffing. Table III Agricultural Technical Personnel Personnel Paid By Theoretical No. Actual No. Commune agronomist Central Govt. 1 1 Commune veterinarian H 1 1 Commune forester " X 1 0 Bee-keeping specialist Commune 1 0 Fish-farming specialist " 1 0 Coffee specialist OCIR 2 2 Extension agent Commune 10 3 Veterinary assistant " 10 2 TOTAL 27 9 (b) Proleits 107. The existence of "projects" in three quarters of the communes in Rwanda (some may even have two to three projects simultaneously) is a major element in the variety of situations found at commune level. The projects are increasingly confined within administrative boundaries, although this was not the case until recently; for instance, initially the Bugesera-Gisaka-Migongo (BGM) Rural Development Project covered only two and a half communes in the Bugesera region. These "projects" temporarily not only relieve the communes of their financial obligations vis-a-vis the agricultural sector and increase their personnel, they also improve rural life by providing extra cash income to farmers, and construct new infrastructure (housing, offices, warehouses, demonstration farms, seed multiplication centers, veterinary dispensaries, stud and breeding stations, etc.). At times, an integrated rural development project may also lead to the establishment of schools, health-care centers, roads, water supply facilities, etc. - 35 - 108. Nevertheless, a "project" is not without its shortcomings, especially with respect to: (i) relations with the local authorities, (ii) sustainability, and (iii) impact on economic agents. First, most projects create an organization which duplicates, or for a time subsumes, the official apparatus. In such cases, the project management team tends to act relatively independently of the local authorities, which soon leads to conflicts. For instance, there are a number of problems associated with the two-tier extension structure set up in the BGM Project zone. The prefecture, commune and 'secteurs" are responsible for execution of the MINAGRI program, while the project management team has its own program to carry out. There arise numerous issues of jurisdiction and responsibility, resulting in short circuits in the normal transmission of information. Confrontation between local authorities and the personnel of "external" project can be counterproductive, especially when the latter are considered as compensating for the lack of local skills or financial resources.46 109. Second, while "projects' provide the means to implement a program within a certain time period they often do not ensure institutional continuity and coverage of recurrent costs for the post-project period. As a result, one project succeeds another (sometimes through another source of financing), as the only way for the commune to sustain their activity. The agricultural projects at Kibuye (PAK) and Kigali-Est and the BGM, for instance, are on to their second or third phase (a total of 25 years of financing for PAK), and yet no alternative long-term strategy of sustainability has been formulated. The Government and the donors are well aware of these issues and new projects (such as Gitarama) try to integrate their structure within the existing institutions with some modifications, if needed (para. 133). 110. Finally, "projects" have sometimes hampered the initiatives of the commune's usual economic agents or interfered with their emergence. For instance, they may replace local businesses as distributors of inputs, or artisans processing the agricultural produce, or the banks in providing credit, etc. However, projects initially assumed such activities because of a lack of initiative on the part of the private sector or sometimes due to the inexistence of private institutions themselves. An increasing number of projects are now promoting private sector involvement in their activities as this may be one of the necessary steps in order to ensure project sustainability. (c) Agricultural Cooperatives 111. Locally based farmers' groups and associations take various forms (for informal groups see paras. 90-94). There are also cooperatives which enjoy official status and receive government support. Initially, under the impetus of development projects or the existence of "paysannats", large-scale cooperatives (with hundreds, even thousands, of members) were created in the 1960s. Following their failure, the Governu int initiated multi-purpose communal cooperatives in the late 1970s. These organizations had some common features: (i) they arose from national instructions from the authorities or in the context of a "project"; (ii) they were quasi-compulsory in nature (e.g. farmers had to be members of cooperatives to be eligible for one of the plots of land available through the "Office du Bugesera-Mayagal (OBM); and (iii) they were managed by salaried individuals appointed by the authorities and not by the members. 46JC. Ageland J.C. Bejannin, Evaluation: le Prolet Bugesera-Gisaka-Migonpo au Rwanda, Republique Frangaise, Ministere de la Cooperation, mai 1986. - 36 - 112. Some of these cooperatives remain active, after having somewhat evolved. In addition, new organizations are appearing, for instance in the zones covered by the former OBM in the Mayaga region or the AIDR in Gisaka. The Government seems to have adopted a more liberal policy with regard to these cooperatives, encouraging initiatives born at the grassroots level and based on real community interests. The members of these cooperatives are now able to formulate their strategy and objectives and to :hoose the manager according to job ability and educational qualifications. The "Centre de Formation et de Recherche Cooperatives" (IWACU) and some NGOs (COOPIBO, in particular) play an important role in the promotion of cooperatives in general and rural cooperatives in particular (para. 230). (d) The role of women47 113. Almost all women in Rwanda's labor force work in agriculture, and about 25 percent of these are heads of farming households. A third of their time is spent on farm work, mainly on food crops (sowing, weeding, harvesting, and processing), and the rest is taken up with the heavy work of carrying water, gathering fuel wood, grinding grain, attending to small livestock, basketmaking for their families, brewing sorghum beer, and the management of their households and caring for their children. 114. The Rwandese woman faces considerable constraints in carrying out these tasks. She has only the most rudimentary tools: the hoe for all farm labor and the grindstone for milling grain. It may take her half an hour to two hours to obtain water. Ensuring the food self-sufficiency of the family is not an easy task for women, because most cash crops (coffee, bananas) are grown by men on the most fertile land, and/or close to the house.48 Since they are very few fenale extension agents or AIVs, and no women are selected as leader farmer (para. 129), extension services fail to reach and assist women farmers in growing food crops. 115. In part to lighten their work load, and more importantty eo obtain funds on their own,49 women have formed associations for a range of activities: mutual farming help, savings, group work on community land, artisanal, marketing, etc. Ten years of sensitization within the communes on the cooperative movement have contributed to these developments. Over 500 associations consisting exclusively of women were identified in 1986, not including mixed associations, in which women may play an active part. 47/The material in this section has been drawn from various documents, particularly: UNIFEM, Rapport d'une Mission de Consultation en Republique Rwandaise sur les Besoins des Groupments Feminins en Milieu Rural et Urbain, 1986; 0. Ubonabenshi, Place de la Femme Rwandaise dans le Divelopgement Rural, 1984; I. Nibakure, "La Place de la Femme Rwandaise dans l'Auto-suffisance Alimentaire", 1986. 48/In order to have the benefit of organic manuring. 49/I. Nibakure, op. cit., notes that various surveys carried out in association with projects indicate that the major aspiration expressed by women is to have the means of earning their own money. - 37 - 116. Despite these encouraging developments, women's participation in associations remains limited. For :instance (in the five communes of the Prefecture of Kibuye), women make up only 20 percent of the 12,000 members of the various associations.50 The main reason cited for such a low participation rate is revealings "Women don't have much time because they have too many family occupations." This indicates how necessary it is to increase women's work productivity in order to provide them with more time for training and participating in associations. The introduction and use of appropriate technology (e.g., water tanks, grain mills, pulping equipment, grain silos, etc.) could be a major step in this direction. 117. In carrying out their activities, many women's associations encounter numerous problems of a material, managerial and operational nature. Once formed, the operations of these associations are hampered by members' illiteracy and low level of technical and managerial tr.ining, and by lack of means of production, credit facilities and product outlets. They are in need of assistance and of informed advice on which public institutions would be best able to answer members' questions. 118. This need for assistance is met only in part. The cooperatives' adviser (a MINIJEUCOOP staff member) assigned to the commune is expected to promote cooperatives and pre-cooperative associations and to provide training for their members. He has some difficulty carrying out this second function, as the multiplicity of such groups has created problems of supervision and follow- up. In fact, it should be now more important to develop programs for the training and supervision of existing groups so that they can become self- managing rather than to expand the sensitization campaign originally designed to increase their number. The CCDFPs could give priority to functional literacy training for women and to providing some degree of supervision and support. This would allow women's associations (which prefer to have assistance from women) to consult female CCDFP instructors and social counselors when needed.51 119. While the commune obviously has an important role to play in promoting women's associations, communal authorities often seem to favor men. Some communes would rather encourage and promote the establishment of mixed associations (instead of exclusively female groups), on the grounds that they are more operational. Most communes prefer to assign communal fields to all- male farmers' groups. Although women's groups could then enter into land leasing arrangements, they are mostly deterred from doing so by the high cost of land rents in densely populated areas. B. The Operations 120. The major activities in the agricultural sector of the commune comprise: (i) execution of national campaigns ("actions d'encadrement"); (ii) extension activities; (iii) provision cf farm services; and (iv) management of land holdings. With respect to the latter, however, it will be seen further (para. 147) that the commune's authority is limited as long as long as the land legislation In force has not been substantially revised. 50/IWACU, op. cit. 51/UNIFEM, op. cit. - 38 - (a) National campaigns (encadrement) 121. The campaigns start off with nation-wide slogans ("mots d'ordre") publicized every year by the Central Government through administrative channels. The personnel in charge of lencadrement" then introduce the themes to the farmers who are required to apply them, particularly through Umuganda. 1982 was anti-erosion year, 1983 was the year of the tree and reforestation, 1984 that of food production, and 1985 that of manure and compost. A new slogan does not necessarily replace a previous one, but receives priority. 122. The anti-erosion campaign is currently being very actively pursued in Rwanda at the local level since anti-erosion terracing is to be completed throughout the country by 1988. Anti-erosion monitoring committees, with technical specialists among their members, have been set up in each prefecture and are under the authority of a government representative (subprefect). Effective erosion control is clearly a top priority for Rwanda's agricultural development and the Government's determination to achieve this objective is commendable. Its implementation, however, needs to be better adapted to agro- ecological conditions. Indeed, some anti-erosion work has been done in non- priority areas, and proper account has not always been taken of the vulnerability to erosion, the nature of the soils, the gradient vegetation cover, etc. One may see anti-erosion strips which do not follow contour lines - - and even sometimes follow the gradient line -- or which have disappeared, the grass cover having been destroyed by straying animals and lack of attention from farmers. 123. The campaign to thin banana plantations is another major action, designed to encourage (food) intercropping on these plots. At times, it has been applied regardless of the land and labor conditions prevailing on a particular plantation. While thinning is totally appropriate on plantations of less than one hectare with abundant manpower, it is not technically or economically feasible on larger plantations where the man/land ratio is low. Furthermore, this is not a well-received campaign by farmers since it is claimed to result in a reduction in the total output of banana beer. 124. The pruning and maintenance of coffee trees is also obviously a matter of major concern since the coffee is the principal foreign exchange earner of the country. The agronomic recommendations are technically sound, and OCIR-Cafe provides pesticides, spray equipment, pruning shears and saws. However, the implementation of maintenance operations (weeding and mulching) again does not make much allowance for variations in conditions from one plantation to another. In fact, coffee growers do not sometimes mulch because of the considerable distance of coffee trees from their housefold and the lack of vegetable material to use for the purpose. 125. The development of compost pits became a 'mot d'ordrel in 1985, designated as the year to increase farm production through compost and farm yard manuring. Each farm was to have at least one covered double-chamber compost pit. The program was put under way without much consideration of local differences, such ass (i) the availability of livestock on the farm; (if) the need for surface mulch; (iii) the appropriateness of compost heaps (instead of compost pits); and (iv) the availability of other techniques utilizing organic matter in ways consistent with established local farming practices and the - 39 - physical location of each farm. The purpose of a compost pit and its method of construction were not always explained to fanmers, and this is why many pits are poorly built (insufficient area/depth, light shade, etc.) and poorly managed (too little vegetable matter incorporated, for instance). 126. On the whole, the way these campaigns are executed reflects: (i) lack of knowledge of actual farming practices (and sometimes unwillingness to recognize their practicability and effectiveness), resulting in non- differentiated application of technical messages; (ii) inadequate research on farming production systems which would provide appropriate solutions; and (iii) a lack of adequate training at all levels. The agricultural professionals, for instance, are trained in a mechanized (non-labor intensive) technology heavily dependent on inputs which, in fact, is much less complex than the practices used by most Rwandese farmers. Apart from inadequate training, there is al 3 a misuse of agricultural professional and assistant level technical staff. The extension agents too often act simply as channels for transmitting orders from higher up. The passing on of specific messages (converted from national slogans) by extension agents acting as communal police officers in no way fosters inventiveness and adaptability among farmers, or allegiance among the instructors themselves. 127. The campaigns launched by the Government are fully justified and consistent with the country's agricultural development priorities, and in many cases, (anti-erosion and reforestation in particular), they have brought undoubted benefits. At the same time, the inflexible and very bureaucratic approach used has had adverse effects, as well as a high cost in human resources (time wasted and loss of enthusiasm among both campaign and extension personnel and farmers themselves). It is difficult to come up with a more specific assessment of the results. Overall, they may be somewhat positive in the short term, but are probably not in the longer term. The campaigns should be executed as a communal undertaking with greater local involvement, probably w.th the help of commune councilors and wresponsables de cellules" instead of extension agents (as in Gitarama project, para. 133). These new 'organizers' could be given the necessary training through the CCDFPs. (b) Extension system 128. The National Extension System (SNV -- "Systeme National de Vulgarisation") was created in the early 1980s to disseminate specific technical messages. Extension committees were then set up at all levels of the administrative hierarchy -- prefecture, commune, and "secteur" -- their role being to reinforce the above-mentioned agricultural campaigns through the extension system in order to reach a greater number of farmers. 129. Extension activities consist of: (i) sensitization and provision of information; (ii) training; (iii) demonstration; and (iv) follow-up. There is generally a sensitization period at the end of any Umuganda session or in the course of sector meetings organized by commune councilors. Training sessions are organized at "secteur" headquarters for pilot farmers on a range of technical subjects. Demonstrations are carried out on the farmers' own land. The communal authorities require extension agents to visit 20 farmers (and their farms) in each 'cellule" every month, and it is this group that constitutes the - 40 - so-called lead farmers (upaysans progressistes"), who are responsible for passing on technical messages to other farmers. They are selected through a competition held annually in each 'secteurl ('Concours Agricole") and designed to reward the farmers who are most successful in applying the technical messages disseminated by communal extension personnel. 130. The National Extension System has met with some resistance, particularly from the aprojects" which wanted to preserve their own extension system (despite no particular technical justification for this stance), but also from MININTER, which had some reservations, as it was setting up its own agricultural training program to be carried out through the CCDFPs. In fact, SNV has been subject to various practical limitations: (i) debasement and rigidification of technical messages as they become transformed into instructions and then into obligations while moving down the administrative hierarchy; (ii) a lack of any real feedback of information reflecting the farmers' viewpoints; (iii) overlapping of and some confusion between extension messages and government campaign activities; (iv) the uniformity of messages, usually directed at the more well-to-do individuals (with large farms, plenty of labor and livestock, etc.) and often selected as leader farmers; (v) lack of concertation between extension agents and farmers on real life farming constraints; and (vi) lack of research on farming systems. 131. The Government is well aware of these shortcomings. A workshop on project reorientation was held in the Spring of 1986, and a national seminar on extension was organized by MINAGRI beginning 1987. Meanwhile, NININTER is reflecting on the CCDFPs (paras. 174-176). The technical structure of SNV is also being re-examined, and alternative modes of organization are being tried (at Gitarama, in particular). Nonetheless, these different reviews have not yet led to the adoption of a national polJcy. This whole institutional process of reflection and self-examination is taking place largely without much input from the communes, and the farmers themselves. 132. The Gitarama Agricultural Project52 was designed, as a pilot project (in its organizational and methodological aspects), to generate results that would be extrapolated, with appropriate adaptations, to the national level after five years of project activity. The following method is to be adopted, implementation ways and means being subject to discussions between farmers and the Project technicians, as follows: (i) in each commune, an area representative of the variety of farming conditions (within its boundaries) is selected according to criteria such as local farm size, existence of livestock, non-farm income, number of trees, labor available, etc.; and (ii) innovations proposed are tested under the prevailing farm conditions, and subjected to farmers' evaluations. The particular farmer on whose land the experiment is conducted accepts responsibility for the work involved, while the Project supplies the essential inputs and technical support. 52/The section on the Gitarama Project benefits from J. Bagiramenshi and C. Bazihisina, Le Cas Gitarama: Demarche et Pratique d'une Seauence de Travail au Rwanda, Groupe de Labeaume, 1985. - 41 - 133. The goal of the Gitarama Project is not to create a new organization, but to support existing institutions (communes, People's Banks, NGOs, etc.) and group initiatives (farmers' associations). The communes involved are expected to enter into a contract with MINAGRI.53 Within this contract, communes accept the principles that extension agents should be polyvalent54 and that extension/training activities for farmers must be separated from both the mass campaigns and the provision of services (see next section). The commune is therefore expected to take responsibility for the following activities, but would operate through commune councilors and not the extension agents: (i) maintenance of the fields for multiplication of forage grass and other products to be directly distributed to the farmers; (ii) establishment of plant nurseries; (ii) all anti-erosion related work; and (iv) maintenance and operations of communal forests. The communes assume full responsibility for managing these activities. In this connection, they are expected to work in conjunction with Project staff to develop an annual work program and budget. 134. With the Gitarama Project, the central authorities have recognized a few new principles as explained above. However, throughout Rwanda there is still some confusion between extension messages ("vulgarisationn) and campaign activity ("encadrement'). This is mainly explained by the lack of agronomic responses to real life constraints and the fact that agricultural staff at various levels have not been so far able to modify their approach to technical problems nor their social relationships with the farmers. Nevertheless, there are some encouraging developments, namely the inclusion of a research and development component in many projects, the beginning of farming systems research at "Institut des Sciences Agronomiques du Rwanda' (ISAR), and the search for partners from the private sector interested in pre- and post- production activities. 135. Some consideration could also be given to the concept that extension services be taken over by the farmers themselves, if support were available on demand.55 Lightly structured entities56 -- co-managed by well-organized farmers' associations, the communes, and such technical agencies as MINAGRI and ISAR -- could be especially created for that purpose. They would make it possible to replace SNV, a system whose shortcomings have become apparent. 531cf. 'Contract Governing Coordination Arrangements Between Communes and the Gitarama Agricultural Project, Implementation Volume, Gitarama Aaricultural Production and Minagri Institutional Development Pro1ect, World Bank, 1986. 54/Extension agents should have the necessary training to replace the "monagris" with technical specialties. This "polyvalence" of the agents should change the focus of extension, which has until now concentrated on separate activities (stockraising, cropgrowing, establishing tree stands, etc.) without much attention being given to the fact that a farm is an organic whole. 55/With such everyday problems as introduction of new varieties and new cropgrowing techniques, but also with efforts to reduce the labor component in farm work, improve transportation, and assist farmers' organizations. 56/In the form of a single service agency for each group of three or four communes, to be staffed by a university-trained agronomist and four or five technicians. - 42 - (c) Support services 136. Direct support services to farmers for crop production and animal husbandry are generally provided by the communal technical officers. For agricultural production, these services consist of production by communal nurseries and sales of fruit and coffee seedlings and anti-erosion pasture grasses. As part of national campaigns, these products are in great demand, and, therefore receive the support of MINAGRI and OCIR-Cafe (which finance temporary workers, waterers, nurserymen, as well as technical visits, seeds and other materials, etc.), and mobilize important manpower resources through Umuganda. Communal technical officers also provide selected seeds, fertilizers and pesticides.57 137. In general, the provision of agricultural inputs to farmers produce little results for a number of reasons: (i) it may not always meet the specific demand from farmers; (ii) the particular goods or products are rarely available at the right time; (iii) their quality is often poor and not adapted to farmers' needs; and (iv) technical know-how on their use is not provided at the same time as the product (particularly true in the case of fertilizers). 138. For livestock production, the following services are provided by communal personnel: (i) inspection of meat sold on local markets, -- although the quality of this control is doubtful due to the poor qualifications of the agents in charge; (ii) animal health services which, except during large-scale vaccination campaigns are not provided regularly given the shortage of communal veterinary personnel, their poor technical qualifications (they are hardly equipped to make correct diagnoses or prescribe adequate treatments), and the Irregular supply of veterinary drugs and other products; and (lit) genetic improvements effected via stud and breeding stations. 139. In general, the number of services provided through the communes should be reduced. Public services such as meat inspection, cadastral mapping, production of coffee and forest tree seedlings, and anti-erosion and pasture plants could remain under the responsibility of the communes. On the other hand, the provision of renumerative services, such as procurement and distribution of fanm inputs, should be entirely handled by the private sector, the farmerst associations and cooperatives. (d) Land Tenure and Land Use Planning 140. Although the distribution of vacant land on e large scale is entirely in the hands of MINAGRI, the communes have the responsibility for allocating small land holdings. In practice, very few communes have enough unused land left for agricultural activities. Most available areas are being replanted with 57IMINAGRI's Selected Seed Service supplies seed for multiplication to hprojects'. In non-project areas, the Service operates only -- ai,d not very actively -- at prefecture level, with its seed warehouses. - 43 - trees, sometimes as "communal reforestation" schemes initiated by projects58 or via the Umuganda work, and at others under a "grouped individual woodland' scheme. As far as the country's marshlands are concerned, it is some years now since the communes could make decisions on their temporary assignment; this is now the responsibility of MINAGRI, which endeavors to keep them as reserve areas. 141. Forests/woodlands are assets of considerable importance to the communes, although very unevenly distributed among them. Some communes possess several hundred hectares of forests with a hign financial return, while others have no unused land stock that would enable them to establish forests. Although the Commune Forest Plans as currently set up should lead to an improvement in the management of Rwanda's woodlands, they are usually designed by technicians who fail to take account of socio-economic realities. For communal forestry plan to be effective, the demand for the different classes of wood (fuelwood, lumber, etc.) should be assessed, and appropriate and cost-effective operating methods be suggested to the communes. 142. In the last resort, management of these assets should be of a commercial type with an accounting system which enables to identify all forestry operations. Currently, these are not always traceable in the communal budgets. Depending on their particular circumstances, and with assistance from the Department of Forests and in line with the regulations of the new Forestry Law, the communes could consider entering into contracts with forest management and production companies, setting up such companies themselves in partnership with neighboring communes (Intercommunal Association for forestry), or creating their own forest management units. 143. Compared to other countries in Africa, Rwanda has a rural population that is relatively homogeneous in its farm size and income distribution. By law, all land in Fwanda belongs to the State. Therefore, in principle, unused and uncultivated land may be usesd by the Government for any purpose. All pasture and grazing land is held under community ownership. Farmers in settlement schemes ("paysannatn) enter into a contract with the Government which grants them the use of a plot which is not subdividable. Cultivated land under usufruct rights may be repossessed by the Government, in which case compensation is paid on the basis of the value of crops, trees planted, type of dwelling, etc. However, this arrangement is specific to the "paysannats", which have not increased in scale for some decades. 144. As for other types of arable land, occupancy and use are governed by a combination of traditional law and modern regulations. Farm size is decreasing rapidly because of the laws of succession which require subdivision among all male heirs. In addition, widespread land purchase and sale practices that actually infringe the law have created a legal vacuum in this sphere. Over half of all farming families are estimated to work less than one hectare of land. 58/In such cases, the legal status of these forests and the rights to their revenues are ill defined and are a matter of contention between the communes and the Department of Forests. The latter has proposed forestry legislation that would make define rights of ownership and formulate operating regulations, but in view of the reservations voiced by communes, the National Development Council is "hesitant" to ratify the proposal. - 44 - The traditional family farm consists of a number of plots scattered over the particular "colline", this dispersion being seriously exacerbated by subdivision among heirs. Land is becoming increasingly a scarce factor of production, and the number of landless agricultural workers continues to increase. 145. At the same time a new group of landholders, owners of five to ten hectares, is appearing. This poses a social problem which could become serious in the future, especially if these large land holders are not farmers themselves. Such is the case in the tea-growing areas, "paysannat" plots adjacent to Kigali (ICYANIA), and on a larger scale through rented land in the GBK zone. From a purely economic standpoint, this trend may justify itself in the long term if it results in the creation of agricultural enterprises that are more productive. In general, however, it would be preferable to encourage other methods of regrouping holdings that have less impact on the land tenure system. 146. It is clear that the fragmentation of traditional farm holdings in Rwanda makes it more difficult to take advantage of a number of urgently needed farming intensification techniques, e.g. anti-erosion measures, hillside irrigation schemes, crop/livestock combinations, farm yard manure, etc. Land law, which previously contributed considerably to promoting the growth of agricultural production, no longer seems in step with current realities. In practice, the Government encourages investments in land (terracing for anti- erosion purposes, measures to improve soil fertility, development of marshlands), which are not taken into consideration in the current legislation. Any reorganization of landholding patterns in areas of long occupation is an extremely complex process.59 Nonetheless, a gradual rationalization of land use patterns is necessary in order to reduce time loss (in getting to one plot from another) and to ensure adequate maintenance of and improvements on landholdings. To this end, 'area development contracts' ("contrat d'amenagement foncier") could be established between farming communities ("cellules' or 'secteurs"), and the communal authorities, with the assistance of a national NGO (such as ADRI) specializing in agricultural development. 147. Land tenure policy needs to be better adapted to present conditions. The current legislation should be reviewed in order to improve land use efficiency, limit certain forms of land accumulation, improve communal revenues, and facilitate access to agricultural credit. The first step would be the undertaking of cadastral surveys which would provide better information on spatial land distribution. Many communes have in fact begun registering landholdings within their boundaries. Before designing a new legislation, however, a thorough study should be undertaken and experiments be conducted in specific areas to train suitable communal personnel in all aspects of cadastral work while improving the procedures used in such mapping and recording operations. 59/The Central Government is well aware of this and has plans to study the matter. - 45 - CHAPTER VI. SOCIAL SECTORS A. Education 148. The communes play as crucial a role in the education of the Rwandese population as they do in agriculture, if not more so. At the commune level, the education system is characterized by: (i) organizational complexity, since there are many interests to be represented; (ii) many actors engaged in the collection and management of financial resources with the risk of lower efficiency and accountability; (iii) financing constraints, despite substantial contributions of the population (in terms of umusanzu, fees and labor); (iii) a search for types of institution to supplement the existing education system; and (iv) the establishment of private secondary schools. (a) Types of Education 149. Except for public secondary education, which goes beyond the frame of the commune, the educational system at the commune level consists of: (i) primary schooling; (ii) post-primary technical instruction, through the CERAI; and (iii) non-formal training, through the CCDFP. The commane also plays a major role in the setting up of private secondary schools, generally on the initiative of parents' associations. 150. Given its rapid population growth rate, Rwanda faces the difficult task of meeting the demand for schooling at both the primary and secondary levels. Access to a public secondary education facility is severely limited. Only 9 percent of the school-age population makes the transition from the primary to the secondary system, the lowest figure anywhere in Africa. Most of the communes have no public secondary school, and distribution of these facilities among them is quite uneven. Access to secondary schooling is even more limited for girls, their enrollment proportion being 7 to 19. The limited number of students accepted for general secondary education from each commune (28, on average) means that the burgomaster may intervene for each candidate within rules and procedures set by the Government. 151. Primary school enrollment varies from one prefecture to another, but on a national scale the rate hovers around 55 percent. According to last census figures (1978/79), disparities among the communes of the same prefecture are not very pronounced. Government directives and regional (rather than communal) features such as terrain (topography), settlement patterns, culture, and religion determine primary school enrollments at the commune level. Although the Government made commendable efforts to increase the primary school enrollment rate among girls, they are still markedly under-represented in the last year of the primary cycle, owing to the cumulative effects of previous low enrollments and a dropout rate higher than boys'. - 46 - 152. The lack of post-primary educational institutions other than a very limited number of secondary schools in the face of heavy demand for further education led the Government to create the CERAIs, "Centres d'enseinement rural et artisanal intagra" (see Section (d) below). Admission to a CERAI is the only possibility open to students completing primary school who wish to continue their studies but are without access to a secondary school.60 In 1984185, each commune had at least one CERAI, which, on average, accepted into its first-year class approximately 25 percent of children finishing eighth grade of primary school. 153. In addition to formal education, each commune is supposed to have a CCFDP, 'Centre communal de davelopRement et de formation vermanente,"61 designed to provide, literacy, technical and political instruction to young and older adults (see Section (e) below). (b) Administration of Education in the Communes 154. Public education is the responsibility of the Ministry of Primary and Secondary Education (MINEPRISEC), and its jurisdiction includes the post-primary rural and artisanal instruction provided by the CERAIs. The MINEPRISEC delegate at commune level is the inspector ("inspecteur de secteur"), who may in some cases be responsible for two communes. Under the authority of the inspector for the prefecture, he supervises primary schools and CERAIs in the academic and administrative areas. 155. At commune level, the most important decision-making institution is the "Conseil de Commune pour l'Enseignement' (CCE -- Commune Council for Education). Chaired by the burgomaster, the CCE comprises the inspector, a representative of each association sponsoring a (private) school in the commune, the principals of the CERAIs, the principals of the primary schools, two communal councilors, and a parents' representative for each "secteur" in the commune. Like the inspector, the CCE is responsible for both primary education and the CERAIs. 156. CCE has many functions with respect to primary and post-primary education: (a) As far as the school infrastructure is concerned, the CCE, besides ensuring the quality of instruction, is in charge of phasinglstaggering of school-system investments (construction, maintenance, etc.) according to the school map drawn up by the Ministry. In practice, the rate of construction is the result partly of instructions issued by the Ministry and partly of the ability of the communes to mobilize the necessary local resources. The school map leaves the CCE only a narrow margin for investment in new school buildings for reducing possible disparities between communes. 601If they are not admitted to a public secondary school, and/or are unable to attend private secondary school as a result of financial or location difficulties. 611The instruction given in the CCDFPs is supervised by MININTER. - 47 - (b) With regard to personnel, the CCE plays a role in its selection and the evaluation of its performance. It draws a list from which school principals are selected and gives the inspector its opinion on the qualifications and assignment of teaching personnel. While this participation in the selection process offers a means for greater decentralization with accountability, it is essential that the inspector preserve his independence of judgment and that the prefectural authorities be not too heavily influenced by local considerations. At the evaluation stage, CCE's involvement in commenting on schoolteacher performance could have some disadvantages. (c) Decisions on primary-school students who are to repeat or be expelled are made by the CCE. This is a major responsibility which, in view of the cost associated with repeaters could lead to significant reallocations of available schooling resources. The list of atudents eligible for admission to the CERAI, drawn by the CERAI teachers' committee, is given final approval by the CCE. This function is a logical consequence of the commune-oriented focus of the CERAIs. But the fact that students who have failed secondary school entrance examination are not eligible to attend the CERAIs, limits the communes' scope of intervention. Moreover, commune officials act as recruiters rather than selectors of CERAIs' students because of the low demand for such schooling. (d) It is also up to the CCE to "program school-system income and expenditure and monitor budget execution." The programming function mainly involves the scheduling of schoolroom and workshop construction and the assessment of the contributions in cash and kind with which the population is expected to supplement the commune's own resources and government grants. The communal administration is directly involved only in managing capital expenditures related to education. Actual deployment of the resources for equipment and maintenance of buildings is, in principle, the responsibility of each school's own Council. 157. The local population is directly involved in the operations of educational establishments. All parents and adults concerned directly with a primary school or CERAI participate in the School Council which make recommendations, but mainly approve financial reports. The school principal is answerable to the Executive Committee.62 In the case of the CERAI, this Committee consists of numerous representatives of the commune, 621The "Comite d'organisation du centre scolaire" (in primary schools) and the "Comite de gestion' (in CERAIs). - 48 - in particular the burgomaster, the cooperative and youth organizers, and the director of the CCDFP. Although this provides further evidence as to the intention of the authorities to give the CERAIs a commune-based orientation, the unwieldy nature of these structures seems to create some confusion on their respective roles and to restrict opportunities for initiative on the part of principals. 158. Other participants in the educational system are also represented in various instances. The Teachers' Committee at each primary school and CERAI makes proposals on student admissions, repeaters, and expulsions. In addition, the CERAI Teachers' Committee is responsible for managing funds earned on any production work. Each CERAI also has its Students' Committee. 159. This plethora of organizations regrouping authorities, various actors and users reflects a legitimate concern to secure the involvement of all those concerned with the functioning and development of education at the commune level. These institutions provide the population with opportunities for taking part in local affairs, and this probably also awakens civic consciousness in the community. While this setup brings undoubted benefits, it may also lead to some fragmentation of efforts and dilution of responsibility. At times, the control each committee exercises on the other may inhibit initiatives instead of fostering them. 160. Although the existing legislations63 do not give the communal administration a strategic role in the formulation and implementation of educational policies, the burgomaster is heavily involved in the development and operations of the educational system within his own commune by virtue of his chairmanship of various councils. In practice, there is hardly any aspect of primary or post-primary education in which the commune, through the CCE and the various committees, is not totally involved. The effectiveness of this decentralization manifests itself in mobilization of significant volume of resources at the commune level. (c) Financing at commune level 161. Rwanda's overall financial effort at the commune level for public education64 could be estimated at about RF 4.7 billion in 1985, equivalent to 3 percent of GDP.Obviously, the Central Government is the largest contributor providing about three quarters of total funding.The Government meets directly the payroll of almost all administrative and teaching personnel (except caretakers/guards), gives financial assistance in the form of grants to the communes for building schools and workshops, and participates in the provision of teaching aids and technical materials (about RF 123 million in 1985, or US$1.80 per student). 63/The Laws of January 25, 1986 and June 29, 1985, and the Presidential Order of October 10, 1985. 64/Primary, rural and integrated education, plus CCDFP. - 49 - 162. The contribution of the communes is rather complex: As an administrative entity, they contribute about 6 percent to the financing, out of their own funds; as an economic unit, they provide the framework within which the population's contributions (about 20 percent of total financing) are collected. They also provide some assistance in kind%65 (a) Communal recurrent expenditures on education represented about 10 percent of a typical ordinary budget over the 1980-85 period. They were mainly for school maintenance and purchases of teaching materials (equivalent to about US$500 annually), and for salaries of five CCDFP instructors. This budget contribution is relatively small for primary schools, rather limited for CERAIs, but substantial and vital for the CCDFPs. (Table IV). (b) School construction, typically accounts for one quarter of the communal 'extraordinary" (capital) budget. The current pace of primary classroom building (8 per annum at an estimated cost of RF 3.6 million) puts a strain on communal finances. The communes occasionally receive government grants for this purpose, but they must supplement them with the proceeds of special levies (and school fees). 163. The population contributes special levies ("umusanzu") and school fees (wminervall). In addition, they make a contribution in kind through Umuganda. The modes of collection, accounting and allocation of these funds vary from commune to commune, and generally leave much to be desired. Following the reform of education and until 1982, umusanzu and minerval were both budgeted as capital receipts. A distinction was made thereafter between the two to reflect the fact that the first one is paid by the population as a whole (as a local tax) while minerval is paid by the pupils'parents only (as a users'fee). As regards umusanzu, it is levied and collected by the commune, and usually recorded as a capital receipt in the "extraordinary" budget. 164. The accounting of "minerval" is much unsatisfactory. Even before 1982, recording of the proceeds of minerval was most irregular. In our sample of 18 communes, only three had consistently recorded this source of revenue from one year to the next; five communes made no mention of the minerval at all, while in other, its order of magnitude varied considerably from one year to another. In addition, four of the 13 communes which treated the proceeds of the minerval as budgetary revenues prior to 1982 continued to record them as such in 1983, although no longer requited to do so. Since 1982, the educational establishment which actually collects the funds is supposed to keep most of them. However, some uncertainty still exists, and there are several cases where the funds were in fact paid to the treasurer of the CCE. The relevant Ministerial Order (dated August 18, 1986) does not simplify the situation, since it entrusts management of 80 65lTransfers of land (for school gardens, for instance), supplies of materials (for maintenance or construction), and provision of means of transportation during construction. - 50 - percent of the proceeds of the minerval to the Council of the establishment that collects it. The remairider must be passed on to the CCE, where it is combined with similar transfers from the other establishments and used to rectify disparities in school development within the commune. This new system, designed to provide intra-commune equalization or cross- subsidization of education expenditure, has so far been put into practice in only a small number of communes. 165. The proceeds of minerval are in general allocated to building new classrooms. In half of the communes visited by the mission, they were used for maintenance of premises, minor repairs, and reconditioning. In a few instances, some school room furniture (desks and benches) may also be financed. School supplies are provided by the pupils'parents. 166. Primary schools and CERAIs have similar financial management methods. In primary schools, the school Executive Committee (para.158) is in charge of "using available funds and materials for the construction, equipping, and maintenance of school buildings." However, there is not much room for maneuver in utilizing these funds because of the cross- subsidization requirement (para.164) and the control exercised by the CCE. As to the CERAIs, the proceeds of minerval are almost always used to purchase the raw materials for students' work. Unlike the primary schools, the CERAIs rarely use minerval to equip, build or maintain facilities, which is a communal responsibility. The collection of fees and their allocation seem to be more firmly under the control of the Executive Committee than is the case with the primary schools. 167. Some CERAIs aim at self-financing through the sale of the agricultural or artisanal products made by their students. But such cases are still rare, and profit margins quite low because of poor product quality and narrowness of local markets. The expansion of education in the communes should create a market for the CERAIs' products. For instance, CERAIs'garment-making sections, which cannot compete with imported used clothing (friperie), could possibly become competitive by specializing in the production of primary school uniforms.66 Mobilization of various resources (from the communes, parents, and sale of products) is necessary if the CERAIs are to have the minimum of teaching materials; only in a few cases is this minimum actually forthcoming. 168. It thus appears that a maximum degree of decentralization, compatible with a satisfactory operation of educational institutions, has been achieved. The direct or indirect involvement of the communal officials and staff in the management of the local education system give the commune an almost complete control over the use of operating funds. It is, however, necessary to more clearly define the respective authority of the school management and the communal administration, as well as the principles for accounting and allocating communal funds. 661This type of experiment is under way in the Commune of Ntongwe. - 51 - Table IV Sources of Financ:ing for Current ExDenditures of CERAIs and CCDFPs Type of expenditure CERAIs CCDFPs Principals Central Govt. Central Govt. Teaching and Training " Commune/Central Personnel Govt.&/ Caretakers/Guards ParentsIcommune!! Materials Minerval/commune/ Commune/external and Teaching Aids self-financing!/ aid Equipment Minerval/communel Commune/external self-fi.nancing aid Maintenance Communei/minervala/ Commune a/ In order of importance. (d) The CERAIs ("Centres d'enseignement rural et artisanal intkre") 169. The CERAIs are typically commune-oriented institutions as they were set up to provide their students with post-primary skills necessary to enter the labor force in the commune itself. In an effort to improve the practical training in the CERAIs and to create local employment opportunities, some communes sponsor, and even occasionally finance cooperatives whose members are CERAIs' students and graduates. Many of these cooperatives (generally woodworking or garment-making workshops) are still at experimental stage (communes of Nidora and Kigarama). Bilateral aid agencies and the NGOs are particularl:; interested in this type of action, that suits their integrated development approach. In his efforts to promote such activities, the burgomaster seems to prefer the support from the NGOs or other bilateral aid doncrs to that from national institutions such as the FDC (Cot=unal Development Fund). This quest for external financing might be more productive if an up-to-date list of appropriate organizations was made availatle by the central authorities to the burgomasters. 170. Although the commune has a certain authority over the operations of the CERAIs (selection of curriculum, for example), it is not in a position to deal with the issues these institutions pose on a national scale. The insufficiently defined purpose of the CERAIs, their *mncertain - 52 - tie-in with a long primary education cycle, their inadequately qualified teaching personnel, and most of all the lack of ready outlets for their new graduates constitute a set of problems the communes are hardly in a position to solve. 171. The CERAIs experience difficulties in attracting students. In four communes visited in mid September 1986, for instance, no more than 15 prospective students had requested admission to the first year. Out of the whole sample, only one commune indicated that it had had no difficulty in filling its 42 first-year student places. Burgomasters pointed to parents' lack of confidence in the quality of the CERAIs (in other words, in the poorly qualified teaching personnel), and their fears that their children would be unable to find employment locally to match the skills they had acquired. 172. In fact, very few CERAI graduates remain in the same commune to engage in artisanal activities. Even cooperatives (when they exist), do not offer any medium-term, let alone long-term, employment security. Some CERAI graduates manage to find work as teachers (under-qualified), policemen (recruited at the same level as primary-school graduates), or soldiers. Others return to a rural life, but probably not with increased productivity. In most instances, however, they leave their communes and make very little use of the skills they acquired during the three years training at a CERAI. It would be, however, imprudent to make an overall Judgment on the effectiveness of the CERAIs based on fragmentary information available at present. It is, therefore, recommended to undertake a thorough study on the employment possibilities and prospects of CERAIs' graduates. (e) The CCDFPs ("Centres communaux de developpement et de formation permanente") 173. These communal institutions of literacy and non-formal training result from the concept of 'an endogenous development based on dissemination by and for the local population of their experience and know- how". CCDFP courses are given for adults and young adults in subjects such as agriculture, livestock farming, basic socio-economic training (hygiene, preventive medicine, civic affairs, etc.), literacy training, artisanal skills, home economics, etc. CCDFPs provide an instruction with both technical and political content at the grassroots level of the population.67 174. The CCDFPs are under the supervision of MININTER and coordinated at national level by an Interministerial Commission for Rural Development (consisting of representatives of several ministries -- MINEPRISEC is not represented -- and of concerned NGOs). At prefecture level, they are supervised by a technical committee whose membership reflects that of the interministerial commission. The CCDFP is actually managed at commune level by an Executive Committee ("Comite de gestion"), chaired by the burgomaster, consisting !.ter alia of the principal and his deputy, commune 67/I. Ndindabahizi, Commune Mugusa: Etude Socio-Economigue; Strategie de Developpement par le CCDFP, Butare prefecture, PPIDC, Mugusa, 1986. - 53 - technical personnel, and commune council members. This Committee is in charge of administering the Center's assets and deciding on the use of its resources. An Educational Committee deals with curriculum related matters. By comparison with the primary schools and the CERAIs, CCDFP is much more under the influence of the communes since ite training is provided in part, by communal technical personnel. 175. A very wide range of personnel is active in the CCDFPs: teachers, nutritionists, literacy instructorst cooperative organizers, and various technical specialists. Some permanent staff are paid via the current budget of the commune (3-10 individuals, on average) and the remainder by the Central Government. In addition, both the commune and the Government make their own technical staff available to the CCDFPs for specific assignments. Finally, there are also volunteers, mainly associated with the literacy campaign and training in farming practices -- either in the context of the Umuganda or by direct arrangement with the principal. Materials, equipment and maintenance of facilities are financed by the commune as and when possible. This often leaves much to be desired, but sometimes external financing (as in the communes of Rubavu, Nyamagabe, and the Kigarama Project) improves the situation. In few cases, the CCDFPs are able to market their own products (as in the commune of Rubavu). (f) Private secondary schools 176. The communes face an uncomfortable situation with respect to post- pririary education. The CERAIs (whose capacity to accommodate students was greatly expanded in recent years), face a stagnant demand, while access to a secondary education is severely restricted despite growing demand. The communes cannot influence the supply of student places in public secondary education, since it is regulated by the central authorities and the "market" is a regional one. A growing private-schools movement (there are already 60 such schools in operation), is likely to alter this situation and to create a less regulated system of supply in which the communes might well find themselves taking a significant role. 177. Since the early 1980s, efforts were made to bring supply of secondary schools more in lire with a pressing social demand for this type of education. Such initiatives are taken mainly by religious groups and parents' associations. The former have been on the scene for a long time already and operate private schools that are officially approved and subsidized, and it is likely that they will expand their presence. The schools set up on the initiative of parents' associations are a much newer phenomenon, however. 178. Parents' associations share certain common featuress (i) they belong clearly to a certain geographic region -- sometimes a commune, or a larger area, but generally not extending beyond a prefecture; (ii) they rely on individuals who are well placed in the political, socio-economic or economic life of the country and who often sponsor the parents' initiatives; (iii) they use land or existing buildings belonging to the Central Government, the commune or the dioceses; and (iv) they collect sizeable funds from their members. - 54 - 179. Directly or indirectly, the commune is involved in these initiatives, generally through the burgomaster, who may act as a catalyst in mobilizing parents to organize and finance their own school. This may be particularly the case within a commune, but it also happens when several neighboring communes combine their efforts. It is common to find the burgomaster, if not at the head (a position frequently occupied by a higher-level official from the Region) then at least holding a key position in the particular association. The burgomasters play a major role in this movement, since they give the associations real presence and power on the local scene. This is particularly important since some parents in the association may no longer reside permanently in the commune they originated and where the school is being established. Naturally, the schools created by parents' associations are intended primarily for pupils who are originally from the particular commune or cLmmunes but they may have not necessarily completed their primary studies there. As well as being a response to a situation of inadequate supply, the wave of newly created private schools also results in maintaining regional cohesion. 180. The private secondary schools that have already started their operations usually follow the same pattern: (i) most started off in borrowed facilities, either provisional or rehabilitated, with the intention of building later; (ii) they have recruited teaching staff with a range of qualification levels -- in most cases there is at least one university graduate with a four-year degree, a number with three-year degrees and others with the secondary school leaving certificate -- all better paid than public school teachers; (iii) pupils are frequently selected by competitive examination; (iv) annual fees are high -- from RF 10,000-35,000 for day pupils whose parents are members of the founding association, up to RF 40,000-65,000 for boarders whose parents are not members of the association; and (v) most offer tertiary-stream subjects such as commerce, bookkeeping, and secretarial skills, as well as training for primary school teachers; technical training is poorly represented. 181. This concentration of effort on tertiary-stream subjects is disquieting, since it indicates a tendency for the parents to choose types of training that are relatively easy and not so costly to provide (minimum of equipment, availability of instructors, low capital and operating costs, etc.) rather than a careful reading of the signals emanating from the labor market. This kind of self-organization of social demand, which has little connection with the needs of employers in the modern sector carries the risk of over-investment and over-production in the long run. The Government should closely monitor the developments in this area and, if needed, provide guidance to these private establishments. B. Public Health 182. The involvement of local authorities in health care activities varies somewhat from one commune to another. In each commune, there is a public health care center -- a CSE ("Centre de sante d'Etat") -- and occasionnaly a government-run hospital. There may also be private "Centres de sante agrees' (CSA) and mission hospitals, which are set up and managed - 55 - by religious missions or NGOs. The 1' tslation only stipulates that 'the commune is responsible for maintaining buildings it owns or uses and for financing the expenditures related to hygiene and public health". It does not, however, specify what the commune's financial duties are with regard to its health care eenters; nor does it define the terms 'hygiene' and "public health." This relative diversity, in combination with the imprecision of the relevant legislation, occasionally leads to functional and financial difficulties. Cost recovery in health services and the supply of pharmaceutical products are two major issues in this context. (a) Financing 183. The Central Government has usually built the public facilities and pays directly the salaries of the core personnel. The Regional Health Office and the communes share the responsibility for financing other operating costs. Specific sharing arrangements vary from one CSE to another. In some cases, it is the commune that pays the cost of gasoline and maintenance products instead of the Regional Health Office; in others, financing from the commune supplements that of other public agencies. In addition, as Table V shows, the communes are also financing the salaries of CSE nutrition unit staff, social service and first aid workers, and caretakersIguards.68 184. The CSEs are without sufficient resources to finance their operations adequately. Until recently, they depended on commune's own funds to supplement government financing of operating expenditures. Patient fees now accrue to the communal budget instead of Central Government's (para. 186). In our sample of 18 communes the share of health expenditures in the ordinary budget varies from 3.7 percent to 6-7 percent, payment for salaries accounting for a major part (61 percent) of these expenditures. Many communes complain of the appreciable deficits in their "health budgets"69 185. The administrator of a CSE, who has responsibility for dealing with day to day operations of his institution has no major role in deciding con the scale and type of its operating expenses. He is limited essentially to pointing out the issues and to formulating demands to the commune authorities. Comments made by CSE managerial staff on communal actions frequently point to the following shortcomings: (i) repairs are either not carried out at all or only after some delay; (ii) ambulances tre not always available, because of breakdowns; and (iii) in most communes, the authorities take no particular steps to improve the supply of pharmaceutical products. In some cases, a burgomaster and a CSE administrator have a good working relationship, but in others they have virtually no contact. It is therefore recommended that more official contacts be established between CSE administrators and communal authorities. One solution might be to set up an Executive Committee in 68/In the case of the 'private" CSAs, the commune only pays the salary of the head of the nutrition unit. 69ISimilarly, the operating expenses of the CSAs are financed by patient fees supplemented by grants from missionary organizations. - 56 - each CSE comprising the burgomaster, the administrator, the commune accountant, and the medical director of the Regional Health Office. This would facilitate the exchange of ideas and information and possibly give the administrator some say in the allocatior. of proceeds from medical fees. Table V Financing the Operating Expenses of the CSEs (Centres de sante d'Etat)l/ Public Agencies Type of Central Health Expense Government OPHAR Region PEV 2/ ONAPO 2/ Commune Salaries: Director * Nurse(s) * Midwives * Nutrition Instructor * * Social worker * * Vaccinator * First aid worker * * Watceman * * Medications * Contraceptives * Vaccines * * Water and Electricity Repairs * Petrol (refrigerator) * * Gasoline (vehicles) * * Office supplies * * Maintenance Products * * 1/ An asterisk denotes some financial responsibility with respect to the CSE. 2/ PEV, "Programme elargi de vaccination" (Expanded Program for Immunization); ONAPO, "Office national de population" (National Population Office). (b) Cost recovery 186. Better cost recovery of health services in the CSEs appears necessary to enable the communes to pursue a dynamic policy in the health sector. Until July 1985, proceeds from patient fees were transferred to the Central Government through the Regional Health Offices. New legislation has since authorized the communes to receive and administer these funds. However, as fees for CSE medical services have not been adjusted since 1974, they are well below those charged by CSAs and the - 57 - centers where traditional medicine is practiced.70 This indicates that the population is willing to pay more for health care than the rates currently billed by the CSEs. From interviews with burgomasters and other officials, it appears clearly that an increase in medical fees would be accepted by the population, particularly if accompanied by an improvement in the availability of pharmaceutical products. A recent study, carried out in four communes, confirms this observation71 (c) Pharmaceutical products 187. In view of the widespread shortage of medications, the Ministry of Public Health and Social Affairs (MINISAPASO) requested each commune to set up its own pharmacy in order to improve supplies. As of September 15, 1986, there were 70 such units. The forms of ownership and methods of management of the pharmacies have been left to the discretion of the communes, but most are privately owned and operated on a commercial basis, with only three communes having set up their own "public' pharmacy. Although creation of these facilities has improved the supply of pharmaceutical products in the communes, their availability to the general population continues to pose a major problem. The percentage of patients unable to buy medications at the local pharmacy is estimated to average 50 percent in the four communes actually visited. On the national scale, distribution and pricing create problems which are beyond the capacity of the communes to handle. 188. The current legislation prohibiting competition in the distribution of pharmaceutical products in Rwanda is one of the main reasons for their high prices. BUFMAR is a NGO which supplies the CSAs with these products, while OPHAR, which is a department of MINISAPASO, is responsible for importing pharmaceutical products and materials, and for storing and distributing the products intended for the CSEs. MINISAPASO also operates a pharmaceuticals production unit, LPR, at Butare, which produces essential medications at competitive prices. Neither BUFMAR nor OPHAR or LPR has the right to sell to private or communal pharmacies. The private wholesalers who are in fact the suppliers of such pharmacies generally import brand-name eroducts at prices well above those charged for comparable generic products.32 70/A series of consultations associated with an episode of illness generally cost RF 20 at CSEs, but RF 50 at CSAs and possibly as much as RF 300-550 at certain traditional medicine centers. 711D. S. Shepard et al, Self-Financing of Health Care at Government Health Centers in Rwanda, Harvard Institute for International Development, April 1986. The percentage of households interviewed that would accept an increase in medical costs, provided it is accompanied by an improvement in access to medications, ranged between 88 percent and 93 percent in the four communes. 72/Rwanda's price control system, which allows a certain profit margin over and above the cost of an imported product, gives private importers no incentive to obtain their supplies from cheaper sources. - 53 - 189. The Government intends to create a new mixed enterprise, SONAPHAR (to replace OPHAR), which will have the right to sell pharmaceutical products to privately owned pharmacies. Although introduction of more competition into the pharmaceuticals distribution system will be welcome, the creation of a mixed enterprise for this specific purpose is not necessarily the best solution. However, before a final conclusion on this point can be reached, a detailed study would be needed on the problems associated with the distribution of pharmaceutical products. - 59 - CHAPTER VII. UMUGANDA 190. The concept of non-remunerated community work, known as Umuganda, was first introduced in February 1974 by the Second Republic. Despite the determination of the political authorities to give a new vigor to this concept and its aims, a large segment of the population saw the idea and its practical implementation in the communes as a return to the forced labor which the colonial authorities -- both civic and religious -- had imposed for decades, until the system was abolished following Independence. In spite of this initial handicap, the authorities were successful in making Umuganda the cornerstone of a strategy aimed at mobilizing "abundant national resources," among which labor ranks first. 191. Umuganda is a national effort undertaken entirely at commune level. Its organizational structure -- from the central authorities all the way down to the commune's constituent "cellules" -- is rather complex. It is designed to involve the entire population in community work. The projects are extremely varied and the people's perception of them reflects this variety. An evaluation of Umuganda therefore calls for considerable nuancing, but some improvements could certainly be made in order to obtain the most effective results from this policy. A. Organization of Umuganda 192. The content of Umuganda is not defined in any legislative or regulatory document. Any attempt to define it would contradict the aim of this form of mass mobilization, which "is to promote awareness and to involve each citizen in the work of national construction." Umuganda is based on a sense of moral and political obligation that the MRND is responsible for fostering. It has three basic aims: (i) political: to mobilize the population in development efforts and to "rehabilitate manual labor," particularly among urban dwellers, with civil servants and managerial staff being expected to set an example; (ii) economics to supplement the country*s limited financial resources by utilizing the abundant supply of labor, and often to provide the counterpart manpower needed at national -- or commune -- level to carry out development projects financed by outside agencies; and (iii) social: to encourage group work and to revive the traditions of mutual aid and support, the concept being "to keep alive and encourage a spirit of brotherhood among superiors and subordinates, thereby setting the stage for an exchange of ideas on specific social issues in the form of a frank dialogue." This aim highlights the desire of top- level government officials to foster dialogue between the citizenry and the Central Government. - 60 - 193. The range of Umuganda activities is extremely vast and can be adapted to any local situation. In the urban setting it might include roadworks, the construction or maintenance of buildings for community use. In rural areas the scope is even greater: cultivation of demonstration plots, reforestation, erosion control, development of fish farming, the upkeep of nurseries, small pilot farms, coffee, banana and cinchona plantations, the construction and maintenance of roads and bridges (particularly local feeder roads), classrooms, commune and sub-prefecture offices, health centers, CCDFPs, and water supply works. According to MININTER estimates (during the period 1974-85) the work undertaken within Umuganda was distributed as follows: 56 percent for erosion control and reforestation, 15 percent for construction of buildings (schools, commune offices, etc.), 21 percent for commune roads, 3 percent for water supply works, and 3 percent for agriculture. 194. Umuganda is an undertaking of national interest that involves each and every citizen. All able-bodied adults are expected to participate in Umuganda; minors, old people and retirees are exempt. The participation rate is considerably lower in practice, however. Each person is expected to do half a day's labor per week. The days earmarked for Umuganda are Tuesday, Thursday and Saturday morning. Most major activities cease during the time allotted for Umuganda. 195. Umuganda's political dimension takes precedence over any other consideration, and this is translated and repeated at each level of involvement and responsibility. Every leader has a key role to play, if only in terms of the example he can set. The MRND's role is decisive -- from its grassroots units ("cellules de base") up to the Central Committee. The Party continues to be responsible for "the philosophical formulation and political orientation" of Umuganda while MINIPLAN has delegated the responsibility for its implementation to MININTER. 196. MININTER is the "executive authority" and it has a special department responsible for performing this role. The fact that this responsibility is assigned to MININTER rather than to a technical ministry, such as the Ministry of Public Works, underscores the priority placed on the political dimension of Umuganda; the recourse to the most structured ministry also demonstrates the desire to formalize its organization and to make it virtually a Government-imposed obligation to be enforced at the various levels. 197. Umuganda is organized and monitored at three levels by special commissions: (i) the national commission, which is chaired by the Secretary General of the MRND and whose members are the political leaders from most ministries directly involved (Planning, Public Works, Interior, etc.) as well as the Secretary General of the Rwandese Chamber of Commerce and Industry - the latter to demonstrate the involvement of the private secttor and the vice-president of the National Development Council; - 61 - (ii) the prefecture commissions, which organize and, above all, monitor the implementation of Umuganda in the communes, and; (iii) the commune commissions, which are chaired by the burgomasters in their capacity as MRND representatives, and which have "some autonomy as to incorporate local initiative and to adapt the program to local characteristics." 198. The primary aim of these commissions is to ensure that the works involved are carefully monitored and evaluated from a socioeconomic and technical standpoint and to pave the way for future discussion on "the adaptation of Umuganda to the needs of local communities." For the past three years, burgomasters and prefectures have therefore had to submit detailed reports, in a standard format, on their activities and have been urged to improve the planning and evaluation of the works. 199. At the commune level, the burgomaster is responsible for organizLng Umuganda. Like tax collection, Umuganda is one of his key instrumionts to ensure implementation of the tasks for which he is respons:Lble. The following table VI summarizes the various phases involved in developing a commune project, for five different types of infrastructuret together with the respective responsibilities of the various institutions involved. Table VI SUMMARY OF COMMUNE OPERATIONS INFRASTRUCTURE AND SOIL PROTECTION SWAMP MANAGEMENT WATER SUPPLY COMMUNICATIONS AND CONSERVATION _ ) CONSTRUCTION IDENTIFICATION Foreign aid ;n Foreign aid or Conmunes Foreign aid or 'Direction Communes most cseso, other- 'Direction O4n6rale' 0&64rale du G4nie Rural' wise by 'Direction des Ponts et Chaus%6es (DG2) O6n6rale de l'Eau' (Dą.PCH) in the case (DOE) or, in excep- of road ulignment or tional cases by civil engineering the commune structures STUDIES Foreign aid or DOE Foreign ald or Communes (preparatory Foreign aid or DOCR Foreign aid or technical DO.PCH work) or projects In ministries support of specific coammnes EXECUTION NUOs, local firms, Umugando and Umuganda (except Peasants supervised by Umuganda and local firms prooters, projects projects In for radical terraces project or by DOGR In the case of communal 0" with usuganda support of specific A projects in support (very limited action) buildings and social involvement communes of specific communes infrastructure. Foreign aid and local firms in the case of rural and social construction, with umuganda involvement. MAINTENANCE User associations umuganda Poasants Not done Umuganda or local and Water Committes umuganda builfing firms - commune water on community and State (*)by drainage and attendant (if there is land irrigation permitting on-), with umuganda year-round farming (dry Involvement and rainy season) - 63 - 200. Umuganda is a good indicator of the capacity of commune's political and technical leaders to undertake the tasks of development in the broadest sense (design, implement and execute a local program or project) which are required of the commune, and to achieve community participation and mobilize available resources. 201. In practice it is up to the commune council to formulate and prepare the program of work with help from its members, the commune councilors who are in a better position to identify concrete needs. It is then these councilors who are supposed to organize the works and see to it, with the assistance of the "responsables de cellules", that they are well executed. In the words of one burgomaster, this organization can be summarized as follows: 'The commune decides, the secteur organizes and the cellule imposes." Each commune councilor and burgomaster is given a large amount of latitude in formulating the program, in deciding on the volume of works and their implementation. This leeway depends, to some extent, on the organizational capacity of the councilors and "responsables de cellule" and on their willingness to supervise and monitor the works, i.e. to expend a large amount of personal energy. B. The Work and the Community 202. The perception of Umuganda varies considerably from one commune to another. In many respects Umuganda reflects the type of communal management and the "style" of the burgomaster. In certain communes the conditions of work are relatively flexible ("I don't want to have any work done that is not really necessary"; "Umuganda must be in the interest of the people"). As a result, the time spent on Umuganda is reduced; exemptions are generous or widespread, particularly where women are concerned; and the volume of work is kept to a minimum. For other burgomasters, however, "The main concern is that people obey, and learn to work for the community"; "Even if they cannot see the point of what they are doing, there is plenty of time for them to understand later." 203. Community motivation also varies widely with the type of project to be undertaken.73 Among the least popular are roadworks which are almost always repetitive, often involve traveling some distance, and are hard to execute when only agricultural implements ere available. This work is therefore seen by the rural population as an endless and tedious task, especially since they themselves make only limited use of such infrastructure. Rural communities do show an interest in intercommunal roads, however, without which it would be impossible to meet certain occasional needs, such as the evacuation of the sick, deliveries close to the home (of corrugated iron sheets and tiles, for instance), etc. 204. The construction of government buildings is also relatively unpopular. A huge labor force has been mobilized for this purpose in the past two to three years, but the task seems nowhere near completion. In this area it is perhaps no exaggeration to say that the population has reached saturation point. After the completion of the national campaign to build additional classrooms in order to implement the elementary education 73/This section draws upon the reports and findings of various earlier surveys and questionnaires on related topics. - 64 - reform, MINISTER launched a vast program for the construction of government buildings (in sub-prefectures and the communes) to be followed by the building of housing for burgomasters and technical staff and, "secteur" offices. These works, which are considered nonproductive and low priority are all the more unpopular because the weekly Umuganda system does not readily lend itself to this type of construction. 205. In the case of the other projects (reforestation; construction of markets, storage sheds, and health centers; development of water supply points and delivery systems), the population's interest depends on their involvement at the project design stage and the benefits they can expect to gain from It.74 But here too attitudes are not always predictable: even in the case of water supply works, whose purpose is eminently social and community oriented, "beneficiaries" can feel pressured or detached if the operation is carried out without prior consultation, if the location of the water supply points seems arbitrary and, particularly, if the arrangements for subsequent takeover are imposed from outside. 206. One of the most favorably received projects is reforestation, despite the need for subsequent maintenance and supplementary planting. In the eyes of the future beneficiaries, reforestation (for soil protection) is the most spectacular rural development operation, because in few years it changes the countryside and can, in certain cases, have a significant effect on erosion, water flow from springs and river flooding. 207. Erosion control calls for special mention here. Minimal personal interest in erosion control goes hand in hand, in discussions among farmers, with unswervingly positive opinions about the work. This can be seen as the result of intensive national campaigns on the subject, to such an extent that nobody would dare to question the soundness of a practice which enjoys such a consensus. Beyond stating the principle ("erosion control protects the soil"), it is hard to get the farmers to be more explicit as to the effects of erosion control on their own farms. Several people see no advantage in having constructed ditches on plots far away from their home, because the grass that is supposed to anchor the soil only barely survives after plarting; and maintenance of the ditches is, in any case, not at all thorough. 208. The complexity of attitudes and perceptions among the farmers reflects the ambiguity that exists in the general attitude of the people towards Umuganda. On the one hand, both the sensitization program and the organization of this unpaid labor are perceived as a major constraint by the population. The extent of this pressure varies considerably by commune and "secteur" but, any relaxing of pressure by the prefectures on the commune leaders and on the commune inhabitants (organization of works and sites, monitoring of progress, control of weekly work quotas) automatically results in high absenteeism or in abandonment.7 On the other hand, 74/"The work is done collectively but not everyone benefits". 75/"People would automatically elect someone as burgomaster who promised not to Impose umuganda" (MININTER). - 65 - whenever the works contribute directly to an improvement in their well- being the people are eager to obtain the results and well aware that there is no other means of achieving them. The paradox of Umuganda is that, despite the initial constraint, it gives communities the opportunity to be directly involved in collective projects and to express their wishes openly, along with their degree of support for the proposed undertaking (or their disapproval by signs of resistance.)' 209. These widely differing views therefore call for a more flexible approach to Umuganda by public authorities. A revealing example is provided in a commune whose burgomaster is. known nationwide for his authoritarian approach and where the comments gathered from the farmers amount to a feeling of working under duress.76 But this commune is also one where the farmers interviewed were particularly satisfied with the performance and coverage of community services: health, nurseries etc., and the existence of a dynamic commercial sector, well-supplied markets, etc. 210. More uncertain for the future, however, is how the central authorities will meet the challenge of getting the civil servants, particularly those of the capital city to perform manual labor with a sense of purpose, "In their case, it must be admitted that Umuganda has no economic justification -- quite the opposite, it costs us money."77 The principle of this work has always been defended in the name of the political strategy of national mobilizatioa and equality among citizens.78 The fact remains, however, "that it is bec.,ming increasingly difficult to keep 10,000 people usefully employed every Saturday in and around Kigali." In this case, plans had to be made for projects or works that mobilize large numbers of people in order to simplify organization, transportation and supervision. But "these must be limited to the communes bordering on the capital in order to avoid spending the whole morning on unnecessary travel and to avoid excessivu fuel costs." C. Evaluation (a) Quantitative Results 211. A strict economic assessment of Unuganda is diffiault in quantitative terms. According to official statistics, Umuganda has probably contributed the equivalent of RF 1-1.5 billion per annum since 76/'It's worse than during colonial times", "Soon we will no longer have time to work on our farms"; etc. 7761Civil servants perform Umuganda during their paid working hours, and it is their ministry's responsibility to get them to work sites. They are also insured against possible accidents on the job, unlike the rural people, among whom a number of very severe cases of accidents on the job have been dragging on for several years, despite appeals by the burgomasters to the Party. 781"We must therefore overcome the short-s:ighted look for an immediate return wnich is prevalent among certain critics of Umuganda (MININTER document). - 66 - 1977, i.e. 5-6 percent of the amount of total investments undertaken under the National Development Plans. The Second Plan (1977-81), provided for Umuganda to contribute 10 percent of total capital investments in the form of labor, i.e. RF 5.5 billion. But "because of inadequate programming, a lack of logistical support and sometimes insufficient mobilization of human resources", the quantitative goals were not met and the actual investment was estimated at only one third of the planned effort. This assessment is in agreement with MININTER's calculations based on commune statements, according to which the annual contribution amounted to about RF 1 billion.79 212. During the Third Plan period (1982-87) MINIPLAN expected that Umuganda investment would represent about RF 10 billion, i.e. 8 percent of the total RF 125 billion in national investments. This expectation was based on the following calculation: 52 half-days at a minimum wage of RF 100, multiplied by 1.15 million households,80 would yield RF 3 billion per annum and RF 15 billicn over five years, of which only one-third would cortespond to maintenance works or intermediate consumption rather than to capital formation. According to MININTER reports, actual investments for 1984 and 1985 were RF 1.29 billion and RF 1.44 billion, respectively. These results, which are well below the Plan projections, are nevertheless quite sizable, given that they are equivalent to about half the government investment budget. 213. While these figures represent a perfectly respectable order of magnitude, certain reservations must be made concerning the accuracy of these assessments. In the first place, the reliability of the commune's reports is questionable, as is indicated by their extreme diversity. The quantitative results reported by some communes are obviously inflated; they would imply a participation rate 1.5 to 2 times higher than the manpower potential established by government censuses for these communes.81 Other burgomasters, on the other hand, "see no reason for covering up the true facts." In one commune, in particular, only 4 percent of the labor capacity is utilized. Similar discrepancies are found at the prefectural level, also. In 1985, Kigali made effective use of 36 percent of its theoretical potential, Gitarama 21 percent, Ruhengezi 12 percent, Gikongoro 13 percent, etc. In addition, the reported figures for the same prefecture vary r siderably from year to year. Between 1984 and 1985, Umuganda dropp _ by half in the prefecture of Gikongoro; in the prefecture of Cyangugu it reportedly increased 61 percent, in Byumba 95 percent and in Kibungo 280 percent. No explanation is given as to the reasons for these changes. 79/For the entire period 1977-83, the figure would be RF 8.1 billion. 80/Based on the assumption which was supposed to be realistic, participation would only amount to one adult per household. 81/These censuses have a margin of error in the order of 20 percent in telation to the extrapolations based on the 1978 census. It is safe to assume, however, that those who did not declare themselves as residents of a particular commune are unable, a fortiori, to take part in Umuganda. - 67 - 214. A further reservation concerns the quality of some of the reported investments. The technical coefficients used to calculate the time spent on the works are consistent but cannot effectively reflect the results achieved (particularly in the case of works on erosion control, earth roads, water supply, for which the calculations lump together projects that are vastly different, ranging from simple repairs to new construction). Despite the large amount of work completed, its effectiveness is not always evident. The communes have neither site foremen nor works superintendents to show people how to maintain the different types of local roads. The implements used for road mainteaance are unsuitable (hoes). Errors inevitably occur, such as grass being pulled out instead of cut, potholes being 'fixed' with mud, stones and grass, earth removed from drains being thrown upstream instead of downstream; certain essential maintenance tasks being omitted (care of drains, culverts and other structures, planting of grass to stabilize soil on slopes, etc.). (b) Prospects 215. During its 12 years in existence, Umuganda has become an integral part of Rwanda's economic and social landscape. Despite the objections and criticisms leveled at it, its principles have not been questioned. The Intention was for Umuganda to "stimulate activity in the rural areas." The results look positive. Whether the job at hand is to restock and maintain forestry sites, carry out routine maintenance of rural earth roads, construct buildings for community use -- regardless of people's perception -- or carry out erosion control activities, it is evident that the communes and their residents have managed to organize and successfully complete the works undertaken. The shortage of technical and financial resources has been met to a great extent by the people and the communes. These labor- intensive operations are within the reach of any self-organized labor collectives, without any assistance or support from outside.82 216. In general, the overall system is carefully organized and is accompanied by very active sensitization and fairly regular monitoring. There are ongoing discussions to deal with any problems encountered but finding genuine solutions is extremely difficult. Technical constraints are not peculiar to Umuganda; they are more generally attributable to the lack of local capability and to the shortage of financial and logistical resources to supplement and thus optimize Umuganda activities. Issues also arise from increasingly noticeable disenchantment among the population due to the inevitable weakening of the concept over the years. Many people would like to see this potential more fully utilized, and steps taken to counteract the declining enthusiasm of the population, without necessarily tightening up enforcement. The problems of Umuganda can be tackled at several levelss institutional, technical, economic and psychological. 82/It is not intended to discuss here projects or works that are jointly organized as part of development projects or those that provide equipment, materials, tools and technical training for the manpower provided by the commune. While their achievements are spectacular, in comparative terms, such a system does not lend itself to replications how are the stocks of tools to be replenished, how should Rhe technicians be paid after the project hos left or in its absence? - 68 - 217. At the Institutional level the Government is making an effort to open up the arena for discussions about the future of Umuganda and 'how Umuganda activities can be adapted to the needs of local communities." For the past three years burgomasters and prefectures have had to submit detailed reports in a standard format on their activities. They have been urged to improve their project planning and evaluation work. The overall system is carefully organized and is accompanied by very active sensitization and regular monitoring. Under the new quarterly works programming system burgomasters should be able to keep their planning on a more realistic basis. At the same time, the communes need to be given greater flexibility and scope for innovation so that they can better adapt Umuganda to local conditions. This would, for example, enable the burgomaster, within the commune, to change the timing of works to be done in the course of a week and to organize the work with the help of teams working in shifts at the rate of two per day.83 218. At the technical level, a marked improvement in labor productivity and efficiency will call for: (i) the stepping up of technical supervision both at the planning level prior to the works and on the work sites themselves ("responsables de cellule" are not necessarily the best site foremen); (ii) an improvement in logistics: the provision of tools, the transportation of materials and manpower. The farm implements furnished by the participants (generally the oldest they have) are not the most suitable for the work, which results in time wasted, and; (iii) the introduction of a minimum amount of on-the-job training (e.g. the basic movements involved in the task at hand) and some technical background would be essential in order to improve the quality of the work and the sustainability of maintenance and development works. 219. The key to better Umuganda management is therefore to agree to place greater emphasis on technical considerations rather than on priorities that are too exclusively political. Similarly, basic and specialized training for the workers and the procurement of appropriate tools -- even if this runs counter to the current policy of equality vis-a- vis the various tasks ("polyvalence egalitaire") -- would make it possible to set precise quantified goals and might pave the way for the "individualization' of specific tasks. 220. Economic incentives could supplement political motivations, both for organizers and participants. Improved technical support and planning should be accompanied by payment for the skills required (technician or team leader) while better work programming implies a more coordinated and flexible organization in terms of time and space. Just as "responsables de cellule" are remunerated (and protest the ridiculously small amount received), a commitment should be made to introduce payment for other 83/A burgomaster gave the example of construction works where builders work eight hours a day and are assisted by a "cellule" of 300 people, who are broken down into teams of 25, with one team working the morning shift and one team working the afternoon shift each day (for a six-day week). - 69 skills needed.84 One cannot, on the one hand, accept the prospect that "the best trained elected commune officials may refuse to perform their tasks, in the future, without payment' while, on the other hand, denouncing craftsmen and skilled workers who 'cannot wait to leave their rural commune" so that they can get paid for their skills and know-how in the city. 221. The motivation of those who are involved in Umuganda, should also be discussed in terms that look beyond the present rigid and uniform system. It was found, from the above review of attitudes towards Uumuganda, that greater attention should be given to the nature and purpose of the community works. The importance of the consultation, sensitization and demonstration processes should never be overlooked or underestimated. The only way to ensure that people's contributions are effective and that the work is done well is by proving to the worker that the task is useful and worthwhile. 222. The original experience with Umuganda could serve as the basis for a virtual self-organization of development, with teams or groups responsible for establishing, in consultation with the beneficiaries and users, the mos1. appropriate procedures for getting the work done, ensuring the quality of the services expected, and scheduling maintenance and supplementary works. Some projeLts could be assigned to institutions that are closer to the communities and more participative in character, such as user associations or cooperatives, which are better at developing initiatives in a more flexible, creative and motivating fashion. Such actions are not obviously in themselves sufficient to accomplish all the goals of Umuganda, but it would help make it less rigid and constraining. 84JFor instance, to municipal laborers workcing on community development operations. - 70 - CHAPTER VIII. NON-GOVERNMENTAL ORGANIZATIONS 223. Non-Governmental Organizations (NGOs) are particularly numerous and active in Rwanda. Their activities are also extremely varied in nature. Along with the Central Government and some projects," they are one of the most important external actors in the social and economic development from below. A. Multiplicity of Organizations 224. There are more than 160 NGOs in Rwanda, approximately 130 of which are active at the commune level throughout the country. Religious organizations have been present in Rwanda for many decades, while voluntary organizations and other NGOs have only become involved in the last twenty years. Their quantitative contribution to the Rwandese economy is quite significant; in 1984, for example, their total contribution was about RF 1.9 billion (US$23 million), which was equivalent to 10 percent of the country's total investment and three-quarters of the Central Government's development budget. While about 80 percent of their activities are sociplly oriented (food aid, health and nutrition centers, education and water supply), NGOs are becoming increasingly involved in directly productive activities. It is estimated that 70 percent of the NGOs active in Rwanda are working to benefit women. Table VII NGO Proiects (1977-83) No. of Amount in Sector Projects Percent RF millions Percent Social Assistance 14 2.5 3,360 43.3 Education & Culture 149 27.0 1,845 23.8 Health & Social Services 154 27.9 1,031 13.3 Water & Power 79 14.3 546 7.0 Rural Development 71 12.9 333 4.3 Economic Infrastructure 32 5.8 312 4.0 Craft & Industry 47 8.5 221 2.9 Miscellaneous 5 0.9 108 1.4 Total 551 100.0 7,756 100.0 Sources MINIPLAN, Document for the 11th NGO Round Table, Kigali, 1985. - 71 - 225. The NGOs that are active in Rwanda can be divided into the following groups:85 (a) Voluntary Organizations, which basically include groups associated with or belonging to (i) small NGOs, with, in many instances, headquarters in Belgium and whose work is very localized; (ii) religious organizations working in most cases at health centers or on parochial projects; and iii) official national organizations, such as SVN (Netherlands), the Peace Corps (United States), OCSD (Quibec, Canada), and AFVP (France). By virtue of its past history and ethical background, each organization has acquired a fairly specialized image. For instance, the French and Dutch volunteer groups are regarded as the preferred partners of the commune as an institution, the former for their 'Programmes de Participation Integree au Developpement Communal' (Programs for Integrated Participation in Commune Development) and the latter for their rural water supply programs. (b) Support and Financial NGOs, which are also extremely diverse in nature: Mi) some, such as CARE INTERNATIONAL and EURO ACTION-ACCORD (EA-A -- a consortium of about 20 NGOs) receive financial support from foreign governments which entrust a portion of their development aid to organizations that have greater flexibility than their own aid agencies; (ii) others obtain funding from religious organizations (OXFAM, CARITAS, etc.), work with or without volunteers on parochiial development projects, and provide volunteers to institutions that are religious in nature; (iii) some financial NOOs established along humanitarian lines, such as national or voluntary NGOs (NOVIB, for example) provide communes with direct assistance; and (iv) twinning arrangements, which, while lacking any institutional framework, may fall into this category of financial NGOs. (c) National ONGs and Associations are fairly recent institutions: (i) A large number of nonprofit associations (ASBLs) were formed several years ago; some of them, such as private schools, which a.o currently on the increase, are primarily private in nature (see paras. 176-180); and (ii) other national NGOs, such as KORA (para. 239), ADRI (para. 230), ITARA, INADES (para. 230), IWACU (para. 230), which offer alternatives to the "all government" approach to development. They receive financial support from abroad and sometimes enter into ad hoc contracts with government departments. The emergence of these NGOs has been an important contribution in the formulation and implementation of communal development projects. These organizations could provide the communes with a new pool of technical personnel according to their respective competence. 85/The list of NGOs cannot be exhaustive and neither can the classification given be very satisfactory. References elsewhere in this chapter to certain NGOs and their activities are merely given as examples. - 72 - (d) Finally, there are several Associations that coordinate the activities of NGOs by sector of activity.96 The representatives of these various groups meet every two months with representatives of the Ministries of Foreign Affairs, Planning, the Interior, Agriculture, and other ministries, depending on the topic to be discussed. These meetings provide an opportunity for an exchange of information and of working methods among experts (expatriate and Rwandese) and also for reporting to the prefectures and ministries on the latest NGO activities. Seminars days are organized on special topics, such as water supply or reforestation. 226. MINIPLAN and MININTER are in charge of coordinating NGO activities: The responsibilities of the NGO Office in MINIPLAN are to (i) ensure that their projects are consistent with the objectives of the National Development Plan; (ii) draw up a list of projects suitable for NGO financing; (iii) exchange information with the NGOs and consult with them through round table discussions, talks and meetings; (iv) evaluate and monitor projects financed by NGOs; and (v) update the information on NGOs operating in the country. In practice these various functions are handled by only two officials who, for want of logistical resources, find themselves unable to totally meet these objectives. MININTER, for its part, is responsible for administering current NGO operations. There is no formal mechanism for coordination between MINIPLAN and MININTER. Apart from the above-mentioned meetings organized by the sectoral agencies, and the round table discussions that take place every five years (1980 and 1985), NGOs have no means of avoiding duplication of effort or expenditure of resources other than to seek out information for themselves on the projects being planned or executed in their area of activity. B. Their Activities 227. The sectors in which the NGOs are active are extremely varied because, by definition, they try to address all aspects of economic and social life. No systematic record is kept of their activities and what follows is merely an attempt to classify them. A possible distinction could be drawn between "social' activities (health, education, training, water supply) and "productive" activities (agricultural and artissnal) but this would be rather artificial. The various actions undertaken will be discussed in this order, no marked distinction being made, however. 86/The main ones are the "Bureau de Formations Medicales Agrees du Rwanda" (BUFMAR), the "Association de Coordination des ONG pour ie Developpement Rural' (ACOR), the "Secretariat National de l'Enseignement Catholique" (SNEC), the "Bureau Episcopal de Developpement" (BED)--which coordinates the activities of the Catholic Church--, the "Conseil Protestant du Rwanda" (CPR)- -which coordinates the activities of the Protestant Church. "Organisation de Volontaires et des Partenaires Rwandais" (OVPR), -- which has Belgian, German, French, Dutch and Canadian volunteer organizations as its members -- and 'Freres des Hommes", and "Compagnong Batisseurs" (COOPIBO). - 73 - 228. In the area of health care, the NGOs primarily manage rural health facilities (health care centers and hospitals) and are grouped under BUFMAR (para. 189). If the burgomaster needs medical staff or assistance with an immunization campaign he can call on BUFMAR, which will contact a local NGO for the help needed. 229. While primary education is entirely under the government authority and organized according to a nation-wide pattern, the specific mode of operation of the "Centres d'Enseignment Rural et Artisanal Integre" (CERATs) and "Centres Communautaires de Developpement et de Formation Permanente' (CCDFPs) is a responsibility of the commune. Their modus tperandi are often inspired by those of other institutions set up by the NGOs (para. 243). Through these latter institutions, the NGOs also maintain important links with post-primary education (CERAI) and adult training (CCDFP). The NGOs also play an important role in providing assistance to secondary schools. 230. NGOs are increasingly aware of the need to improve the training of members of local groups: (i) the "Centre de Formation et de Recherche Cooperative du Rwanda' (IWACU) was established in 1980, with funding from Switzerland and USAID, with a view to encouraging the formation of associations and providing cooperative members with training in management, accounting and research; (ii) similarly, "Action pour le Developpement Rural Integre" (ADRI) was founded by a Rwandese national to' encourage the formation of local groups and to help them design projects, negotiate loans and comply with government procedures; it thus offers training, management services and legal advice and assists them in identifying sources of financing; (iii) INADES-Formation-Rwanda organizes one-day training sessions devoted to agriculture and agricultural extension; (iv) TECHNOSERVE provides free training and technical assistance in planning, management, accounting and marketing to private firms in the rural sector; and (v) the COOPIBOs have set up a "Centre des Services Cooperatifs' (CSC) in the prefecture of Gitarama in conjunction with IWACU and the Ministry of Agriculture. CSC teaches the principles of cooperative organization and as well as simple accounting methods, programming and planning; it plays a coordinating role by establishing contact with various cooperetives and encouraging joint action. It also finances new investments by pre-cooperative groups through a "Caisse d'Innovation" (Innovation Fund) and help them obtain loans via a "Caisse de Garantie" managed by the "Banque Populaires". In 1985 nine cooperatives requested aid from CSC.8' 231. In the area of water supply two major NGOs operate in the communes. First, the Netherlands Volunteer Service (SNV) operates in about thirty communes. They sensitize the people to the need to pay in order to have clean water. They feel that if water supply projects are undertaken without prior sensitization--which can take as much as five years, the users will not feel responsible for the upkeep of the water points, leading to the breakdown of the system. SNV has also trained local entrepreneurs in cite development and in the maintenance of water points. B7/CSC conducted a survey among 61 of the 150 cooperatives in the Prefecture of Gitarama in 1985 to identify the problems encountered by those wishing to set up cooperatives. The findings of this survey are expected to be available some time in 1987. - 74 - 232. Second, the "Compagnons Fontainiers Rwandais" (COFORWA) were founded in 1972 to create a drinking water distribution network in the Commune of Nyakabanda. COFORWA has gradually extended its operations into other communes, and with the financial assistance of the ministries involved and other NGOs has developed more than 100 water supply networks. They limit their involvement to works that cannot be carried out by local people and provide the communes with technical assistance and the physical resources needed. Before undertaking the works the COFORWAs gather all the future beneficiaries or users and draw up the list of works for which the community (in the form of Umuganda) and COFORWAs will respectively be responsible, together with an identification of the materials that are to be supplied by each of the parties. The users then form an association, select their delegates and appoint a board to serve as the interlocutor with the COFORWAs and sign an agreement with them for the development and the maintenance of the water points. The COFORWAs then organize numerous meetings to get the users to understand that they will be responsible for the network, they must share numerous tasks (surveillance of faucets, cleaning of area around standpipes, repairs, equipment maintenance) and see to it that operating costs are covered. 233. Technical assistance to communes is not only geared to helping the commune administration prepare productive projects but also to improving the training of members of local groups. Such assistance is especially necessary because some lines of credit granted to the Commune Development Fu 4 (FDC) 8 are underutilized due to insufficient project preparation. Some NGOs would be in favor of restructuring the FDC so that it can finance studies. CCCE would, for its part, consider opening a special line of credit and financing the setting up of a small office responsible for advising communes and referring them to consulting firms. In consultation with the Prefecture of Byumba, SNV is investigating the possibWity of establishing a Fund for a small group of qualified Rwandese experts who would be responsible for conducting water supply studies. Initially supported by foreign financing, this Fund would be expected to become financially autonomous. 234. There are some fifty "accords de iumelage" (twinning agreements), coordinated by MINIPLAN, between Rwandese communes and municipalities in other countries, mostly in Germany, Belgium and France. These activities usually begin informally at the instigation of a missionary, diplomat or "cooperant." They result in the establishment of cultural and social relations between the twinned towns and in the financing by European municipalities of health/nutrition centers, water supply or bridge construction projects or the provision of medical supplies or school equipment. The most important twinning agreement is the one that has been in existence since 1980 between Rwanda and the State of Rhineland-Palatinate in the Federal Republic of Germany. Under this agreement 18 Rwandese communes are twinned with 14 German towns and receive assistance for the development of primary education, health, road infrastructure, water supply and artisanal activities. 235. After being active for so long in the social sectors, the NGOs are now becoming increasingly interested in agricultural and artisanal activities in the rural communes. Some examples of this involvement follows: (i) as part of efforts to develop bank credit in the rural regions, some NGOs are looking into 88JSuch as the FF 4 million granted by FAC. - 75 - various ways of encouraging local initiatives; (ii) the "Association pour le Developpement Communautaire de Kanage" (ADECOK) has helped to build a self- financing hospital in the Commune of Murunda; the power generator given to this hospital is used for carpentry and welding activities and for brewing banana beer and, thanks to the profits from these cottage industries, the hospital is now self-financing; (Mii) in collaboration with UNDP and UNCDF, FAC is sponsoring a project whereby carpenters are to be trained in the communes to make furniture for primary schools; the funds needed to purchase the materials and .o pay the laborers will be provided partly by FAC and partly by the Parents' associations who will borrow from the 'Banques Populaires"; the carpenters will make enough desks and chairs for about half the commune schools included in the project; (iv) in the Commune of Niakisu, "Freres des Hommes" is also trainiag carpenters and supplying them with materials and tools by way of an advance, that is to be repaid by the artisan as soon as he sells his first pieces of fuxrniture. 236. To support the Government's effort to promote commune-level stockfarminx, the Catholic Relief Services recently set up a Livestock Bank. A farmer having at least ono hectare of land can borrow a cow plus RF 5,000 on the basis of a contract, whereby he undertakes to raise the animal properly and to repay his 'loan" by giving the Bank the first calf born during the following year. In 1985 -- the Bank's first year in existence -- more than 250 families obtained such loans and most duly repaid their loan. Thanks to this system the peasants have been able to manure and increase yields of their bean fields. Families with less than one hectare can borrow goats or rabbits. 237. Some NGOs provide basic assistance for farming and artisanal activities by establishing Commune Funds: (i) Catholflc Relief Services plans to establish such a fund in Kigoma; whereby the commune will be able to lend for what it considers as priority projects. An expert will assist the farmers in the preparation of project briefs; (ii) UNDP and ACOR plan to establish a revolving fund which ctmunnes would use to lend to groups for productive activities and to finance experts to help them find markets for their products; (iii) EA-A envisages to encourage tanning and carpentry works and also the manufacture of bricks and implements, with a view to stimulating non-agricultural employment. 238. Finally, urban artisans in the informal sector in the Prefecture of Kigali have, since 1978, been receiving a new type of assistance: uJeunesse Ouvriere Catholique" (JOC-Kigali) came to the aid of disadvantaged groups of the populationt cffering training and advice to young people and to independent workers in search of stable employment in the capital. JOC also helped owners of handcarts to organize themselves; as a result they have begun to group together to negotiate with the burgomaster of Kigali in order to obtain temporary work permits.89 Similarly, car washers, metal workers, shoemakers, started to form groups and, in 1982 were officially granted the status of 89/Without a work permit it is not possible to obtain a residence permit; but in order to obtain the latter, proof of having a job is required. As a result, urban migrants who want to work but do not have the necessary piece of paper live in constant fear of the police and confiscation of their belongings; they have neither permanent residence nor a permanent Job. - 76 - professional artisan. A project carried out by ILO and funded by Switzerland helped further the negotiations of such groups with the urban commune of Nyarurenge with a view to obt&ining temporary work permits and loans for the purchase of sites and equipmen.. to operate permanent workshops. 239. In 1985, 40 groups and eight federations of artisans joined forces under the name of KORA ("to workl"). KORA offers artisans various facilities, such as training, advisory services on organization and public relations and it provides them with bank and professional guarantees. A Banque Populaire branch recently launched a program of special loans for KORA members whose representatives sit on the bank's management board. KORA now has 1,600 members in some sixty groups, about five of which include women. KORA is looking into ways of encouraging women to form artisanal cooperatives -- particularly for dressmaking, soap making and basketweaving. Every woman with a job can open an account with the "KORA' in "Banciue Populaire" without her husband's signature. C. Their Role 240. This overview of the NGOs and their activities in Rwanda is only indicative of the diversity of their involvement at the commune level. Given the experience they have gained in the country, their dedication and the time they are prepared to spend on accomplishing their goals, the NGOs are particularly well placed to identify the problems that exist at the grassroots level and to suggest appropriate solutions: (i) NGO projects have a direct impact on the lives of the rural people because they focus on the needs identified by the target groups themselves; these needs are often determined in conjunction with commune leaders and the burgomaster, with whom NGO staff work on a daily basis; (ii) their activities are generally tailored to the community and they respect the traditional customs of these communities; (iii) their actions are generally modest in scope and make use of appropriate low-cost methods; and (iv) NGOs carry out projecte and programs based more on the development of human resources rather than on the execution of major infrastructure works. 241. The NGOs are the preferred partners of the communes in the country's economic and social development efforts. They are more flexible than the large aid agencies; they are also more willing and better able to adapt their projects to the dictates of circumstances. Thus, recognizing the usefulness of NGO action in Rwanda, the World Bank first financed a training program to be organized by IWACU and the COOPIBOs as part of the Gitarama/MINAGRI project (1986) and has more recently worked with SNV, CARE and COPORWA on the water supply works in the Lava Region planned as part of the Second Water Supply Project. 242. There is still room for improvement in many areas of NGO activities, however, on both the Government's side and that of the NGOs themselves. First, the need for better coordination at the Central Government level has already been mentioned (para. 226); this is important for ensuring a consistent approach to sectoral policies throughout the country as well as an aDpropriate distribution of the aid provided by the NGOs among the various communes. One wonders, in fact, whether the flexibility of NGO operations, in terms of the choice of partners, might, despite being one of the NGOs' chief assets, not be a - 77 - source of inequality among communes. It is common for information on "dynamic" burgomasters and "good" communes to be exchanged between "cooperants" and NGOs. This can result in an excessive number of activities -- for which the commune authorities are directly or indirectly responsible. This is reflected in the complaints often made by the NGOs: "The burgomaster is overloaded with work and is confronted every day by the need to meet a growing number of responsibilities with fewer resources" or "There are often too many activities going on and few trained personnel to direct them." 243. Graater attention should also be paid to the economic aspects of the operations. The primary aim must be to ensure the sustainability, replicability and self-supportability of projects. A classic case in point is the activities of volunteers which sometimes lack continuity. Even in a well structured undertaking such as the cooperative movement of Nyakabar.da (which set the model for the creation of CCDFPs) the great dependence of the movement on a religious leader has led to difficulties for successfully replicating the institutions established (such as CCDFP). Another aspect is that NGOs, whose projects tend to have a social bias, do not usually put much emphasis on economic profitability. If an NGO opens a line of credit, for example, this needs to be administered according to criteria that ensure satisfactory cost recovery. A brief financial analysis should be undertaken prior to setting the interest rate of loans to farmers, together with a rapid survey of the demand for the product or activity being financed. Lastly, more caution should be excrcised in selecting the volunteers because some of them are too young, too inexperienced and insufficiently qualified for the responsibilities assigned to them. 244. It becomes increasingly important in a decentralized development strategy, that NGOs and the commune leadership focus their activities on local groups and encourage private sector initiatives. These relatively new actors on the development scene, independent of government agencies and traditional donors, represent an important element for development with the capacity to operate at low cost and to adapt to the needs of the people. At the same time, large donors should make greater efforts to take better account of NGO actions and interventions and determine ways to work closely with them without either partner losing its own efficiency and infringing on its agreements with the Government. - 78 - Annex I LIST OF THE COMMUNES VISITED Prefecture of Kigali s Rutongo Bicumbi Gashora Prefecture of Butare I Ndora Ngoma Nyaruhengeri Prefecture of Ruhengeri: Cyabingo Kidaho Kigombe Kinigi Prefecture of Gisenyi : Rubavu Kibirira Prefecture of Gitarama : Ntongwe Nyakabanda Prefecture of Kibungo : Kabarondo Kigarama Prefecture Gikongoro s Nyamagabe Rwamiko Prefecture Byumba Murambi - 79 - Annex II Table I Composition of Current Revenues (1980-85) (Percent) Revenues General Local from Communal Taxes Duties Other Assets RURAL COMMUNES Bicumbi 8.3 39.5 31.7 2.6 Ntongve 2.1 57.7 26.8 13.5 Kigarama 6.0 44.4 21.8 27.7 Bugarame 1.8 33.3 52.6 12.3 Ndora 23.1 34.2 21.6 21.1 Karama 5.0 56.8 18.5 19.7 Buyoga 8.8 38.3 50.1 2.8 Yinyami 15.1 43.5 15.9 25.5 Rutongo 9.8 29.0 34.6 26.6 Kidaho 5.9 30.1 42.2 21.7 Cyabingo 7.1 58.5 19.6 14.8 Sample average 8.5 42.3 30.5 18.8 URBAN COMMUNES Kigombe 9.4 22.2 55.1 13.3 Nyarugenge 1.8 4.0 78.0 16.2 Nyamagabe 6.6 49.5 30.1 13.8 Ngoma 19.8 18.5 42.3 19.4 Kamembe 6.1 21.9 58.2 13.8 Birenga 4.5 28.9 37.2 29.4 Nyamabuye 7.1 47.8 38.8 6.4 Sample average 7.9 27.5 48.5 16.0 Source: MININTER and mission estimates. - 80 - Annex II Table II Composition and Growth of Local Duties (percent) (1980-1985) Rural Communes Urban Communes Share in Annual Growth Share in Annual Growth Local Duties Rate Local Duties Rate Market Taxes 62 9.9 59 18.5 Taxes on Activities 28 13.2 20 24.1 Urban Taxes 1 - 7 70.3 Other Taxes 9 26.2 14 26.2 Total/Average 100 12.5 100 22.8 Source: MININTER and mission estimates. - 81 - Annex II Table III Composition of Capital Expenditures (Percent) (1980-1985) Purchase of Construction Construction Construction Other Equipment of Administr. of schools of Health Infrastr. Buildings Centers Rural Communes Bicumbi 3 26 45 21 4 Niongwe 3 87 13 - 1 Kigarama 3 28 47 - 23 Cyabingo 5 44 42 8 Kidaho 12 4 54 8 22 Rutongo 20 37 19 7 17 Kinyami 6 39 13 23 19 Buyoga - 87 10 - 3 Karama 2 15 9 - 74 Ndora - 37 55 - 7 Bugarama - 51 39 - 19 Sample average 5 41 31 7 16 Urban Communes Nyarugenge 22 30 32 - 16 Kigombe 41 21 - 37 Nyamabuye 33 33 - - 33 Birenga 19 49 20 6 5 Kanembe 3 - 54 4 39 Ngoma 3 97 - - - Nyamaga'l,e 1 17 76 - 6 Sample average 12 38 29 1 20 Source: MININTER and mission estimates. - 82 - Annex II Table IV Composition of Current Expenditures (percent) (1980-1985) Admin. Security Maint. Agric. Livestock Forestry Education Other Rural Communes Bicumbi 24 14 25 5 2 1 6 23 Ntongwe 30 ' 16 13 14 5 - 9 13 Kigarama 35 22 16 13 1 1 8 4 Cyabingo 38 19 15 10 2 - 11 5 Kidaho 28 16 24 9 2 - 13 8 Rutongo 25 18 16 8 2 3 20 8 Kinyami 28 13 25 11 - 1 11 11 Buyoga 31 16 13 12 5 5 12 6 Karama 31 22 14 10 3 - 16 4 Ndora 32 15 24 8 2 - 8 11 Bugarama 39 22 14 7 - - 12 6 Sample av. 31 18 18 10 2 1 11 9 Urban 'oumunes Nyarugenge 19 15 30 2 - - 9 25 Kigombe 24 19 21 8 4 2 11 11 Nyamabuye 30 19 25 6 2 - 12 6 Birenga 24 19 23 9 - - 13 12 Kamembe 31 22 21 8 1 - 7 10 Ngoma 27 12 25 8 1 2 5 20 Nyamagabe 35 24 15 9 3 1 10 3 Sample av. 27 19 23 7 2 1 10 12 Source: MININTER and mission estimates. - 83 - ANNEX III BIBLIOGTAPHY Agel C. & Bejannin J.C. Evaluations le projet Bugesera-Gisaka Migonao au Rwanda, Republique francaise, Ministere de la cooperation, 1986. AFVP, "Special: Rwanda", Revue de 1'Association francaise des volontaires du progres, No.42, juillet 1985. Bagiramenshi J. et Bazihizina C., Le Cas Gitarama, Groupe de Labeaume, 1985. Banque mondiale, "Contrat regissant les relations de collaboration entre lee communes et le projet agricole de Gitarama", Implementation Volume, Gitarama A Production and MINAGRI Instituticnal Development Proiect, Washington D.C. 1986. Barampama A., A ltecoute des artisans du secteur non structure de Kigalit Ris;ltats d'une recherche-action participative, BIT, Geneve, 1986. Bosomingera A., "La participation de la population a l'administration et a ltexacution des missions de developpement des collectivitis locales au Rwanda", in Les collectivites locales et le pouvoir central, Editions Nathan, Paris, 1984. Bryla C.A., "Rapport d'une mission de consultation en Ripublique rwandaise sur les besoins ess groupements feminins en milieu rural et urbain", 2 juin - 11 juillet 1986, UNIFEM. Cochrane G., Policies for Str.nathenint Local Government in Developing Countries, World Bank Staff Working Papers, No.582, 1983. Correse A. & Temelle J.P., Le ceveloppement communal: Exitences et perspectives, Institut de recherches et d'application des mithodes de developpement, 1985. Delpiano P.G, Les disparitis rigionales en matiere dvicuipement des communes, ROC International, 1981. Derrier, J.F., "Public Works Programmes in Rwandat Conditions for Popular Participation", International Labour Review, Vol. 124, No.5 September-October 1985. Dialogue, "BUFMARs 10 ans de dynamisme", Dialogue, mai-juin 1986. - 84 - Dose R. & Mars Z. ed., Community Development - Comparative Case Studies in India, the Republic of Korea, Mexico and Tanzania, UNESCO, 1981. Fernier L., "Etude de base, Projet d'enseignement agricole", Rapport de mission, USAID, 1984. Fortman L., Peasants, Officials and Participation in Rural Tanzania: Experience with Villagization and Decentralization, Rural Development Committee, Cornell University, *980 Godding J.P., "La deuxieme Table Ronde des ONG", Dialogue, No. 113 novembre-decembre 1985. Gow D. et al., The Rwanda Social and Institutional Profile, Vol.1, Context, Crosscutting; Issues and Recommendations, Development Jilternatives Inc., 1986. Guichaoua A. ed.."L'Afrique des Grands Lacs", Revue Tiers-Monde, Tome XXVII, No.106, avril-juin 1986, Presses universitaires de Fr^nce. Heller A., Les lournees de creativite, Groupe de Labeaume, 1985. INADES, Inventaire ONG 1985, Kigali, 1985. Jemas Y. & Hoben S., The Rwanda Social and Institutional Profile, Vol.3, The Demographic Imperative and Educational Reform, Development Alternatives, Inc., 1986. Laboratoire pharmaceutique du Rwanda, Etude de rentabilite et d'autonomie financiere, Butare, mai 1986. Lemarchand R., Rwanda and Burundi, Praeger Publishers, 1983. Marsden D. & Cakley P., Approaches to Participation in Rural Development, ILO, Geneva, 1954. MININTER, "Pour une approche concrete du developpement ^ommunal", document interieur, 1981. MININTER, Rapport du Colloque national sur la formation permanente orzanise a Kigali du 20 au 24 mai 1986, 2 vol., 1986. MINIPLAN, "IIeme Plan d'action communale, Finances Commxunales", Document preparatoire, juin 1983. MINIPLAN, Developpement communal (Prealables i la traduction et mise en oeuvre au niveau de la commune, avec la participation active de l'echelon communal, des lignes de programme du IIIeme Plan), 1983. - 85 - MINIPLAN, "Esquisse de projet du IIeme Plan d'action communal" document interieur, novembre 1984. MINIPLAN, Documents de la llime Teble Ronde des ONG, 1985. Ministare des relations exterieures, Sous-dl.rection des etudes du developpement, Mithode de mobilisation de l'epargne rurale dans les Pays africains, Republique fran9aise, No.58, decembre 1983. Mores E. & Gew D., Implementing Rural Development Prolects Lessons from AID and World Bank Experiences, A Westview Replica Edition, 1985. Musabimana, Les Interrelations entre Banques Populaires et Tontines (ces de la Commune de Kivuye-Byumba), IWACU, 1986. Musabimana J.M.V. & Tjoelker K., Les organisations svontanees en milieu paysan, Une etude menee dans la zone du Projet agricole de Kibuye, IWACU, 1985. Ndindabahizi I., Commune Mugusa-Etude socio-4conomique d'ensemble - Diagnostic gineral de base, Prefecture de Butare, PPIDC-Mugusa, avril 1986. Ndindabahizi I., Commune Muzusa - Etude socio-economi ue d'ensemble - Strategie de developpement par le CCDFP, Prefecture de Butare, PPIDC-Mugusa, 1986. Nibakure I., La place de la femme rwandaise dans lauto-suffisance alimentaire, 1986. Ntexilylmana A., "Projet de recherche operationnelle - Un essai de st.ategie de financement de soins avec la participation de la population", MINISAPASO, mars 1985. OMS, Etude de faisabilite pour la creation d'une societe d'acquisition de medicaments essentiels au Rwanda, Kigali, 1985. Ralston L. et al., Voluntary Efforts in Decentralized Management, Institute of International Studies, University of California, Berkeley, 1983. Rubina D. et al. Le rendez-vous des femes a Nairobi et apr4s le Forum ONG, 10-19 Juillet 1985, INADES, 1986. Sack R., 'La situation de l'ERAI: Analyse, diagnostic et recommendations", Creative Associates, Prolet d'ensei2nement agricole/USAID, Washington D.C., 1986. - 86 - Shepard D.S et al., Self-Financing of Health Care at Government HealthCenters in Rwanda, International Development Institute, Harvard, April 1986 Ubonabenshi 0., Place de la feme rwandaise dans le developyement rural, 1984. UNIFEM, 'Rapport d'une mission de r.onsultation en Republique rwandaise sur lee besoins fiminins en milieu rural et urbainu, 1986. Vengroff R., Development Administration at the Local Level: the Case of Zaire, Foreign and Comparative Studies/African Series XL, Syracuse University, 1983. Vianney M. & Klias T., Les organisations spontanies en milieu paysan, Une etude menie dans la zone du projet agricole de Kibuye, IWACU, 1985. Voyer M., Prolet d'apvui au secteur non structure de Kigali (Analyse de l'impact dt'une experience participative en milieu urbain, BIT, Geneve 1986. Wilcock D. & Ndoreyaho V., The Rwanda Social and Institutional Profile, Vol.2, Agriculture at a Watershed, Development Alternatives Inc., 1986. NOTES NOTES I~~~~~~~~~~~~N1C NOTES NOTES MAP SECTION RWANDA / POPULATION DENSITY BY COMMUNE, 1978 ? IES DENSIrES DE POPULATION PAR COMMNE, 1978 ( RaIshuru POPULATION DENSITY (PERSONS PER KM2) IHABITANVrS PAR KMh4; L I C - 100 LJ -2. ZAIRE __201 25(7 BIRUNGA NATIONAL PARK | 251 - 320 RC INA 710VAL OTS BIRUANGA 'X. MAIN ROADS ROUTf-s iPR, /IPA WLES RiVERS F/EU VES N NATIONAL CAPITAL -AP .' T I/ PREF'CTURE CAPITAL CHff - iFUI ODF PRfEFCTUREF COMMUNE BOUNDARIES / IM/TES OES c Q4muA'UEs PREFECTURE BOUNDARIES 1/MITESf DE S PR~EFECrURES NATIONAL PARK BOUNDARIE- i 4'117FS DES PARCS NVATIOAIAUA INTERNATIONAL BOUNDARIES \ I Pvepa'00 pa NZes sc AN: ae so e )-I a LI r ,ff, Cane es' u'OSnefJst.-'o a (3Or-' la *cla duappo" f o'o ofe s reye've a I sage of'r eat 'am- ,Ze, es a', el M1 Z A 1,. vM#LAWde la pa' saoot o,anda!e el dfefa Sc;, ofo 0,a~r,e,e3 ofe-ahfo "a,e0.:,. ~ sq.toeo m000030' 'SUe 'es a,SI,fof r,O ecco0,omass os accepteni yes frorfe's ADJ KENYA I -v__ TANZANqiA Da( eso,~ 'Solaam )ZAMBIA BURLJND ---- ( e /~ MOZAMBIQUE 29' IBRD 18479 30~, ) ,! U ~~~~~A N LD Kagi A umbo \ ) > . \ / / ,, ,,) \ T ~~~~~~~~~~~~~~~~~~~~~~A N Z A NI A \~~~~~ a \ ( S 6N~~~~~~~~~~~~~~~~~yagator, VAL~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ETME IP84UNA 1