ES:GCP/RAF/284/NET

Terminal Report

FAO/GOVERNMENT COOPERATIVE PROGRAMME

DEVELOPMENT OF A REGIONAL FOOD SECURITY AND NUTRITION INFORMATION SYSTEM

PROJECT FINDINGS AND RECOMMENDATIONS

SOUTHERN AFRICAN DEVELOPMENT COMMUNITY

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS

ROME, 1998

LIST OF ABBREVIATIONS

1. INTRODUCTION

1.1 PROJECT BACKGROUND

The Southern African Region has a total land area of approximately 68 000 km², with a current estimated population of around 131 million people. As a result of rapid population growth, the demand for food, particularly cereals, is increasing. In recent decades, the region has been hit by a number of calamities which have seriously compromised the food security of its population. Apart from recurrent droughts and conflicts, however, there has also been a process of environmental degradation, putting pressure on agricultural production.

Although there is evidence that the region can produce enough food for its population, food production, especially of cereals, continues to vary considerably from year to year, principally because of the weather, but also as a result of variations in the areas planted and the availability of inputs, such as seeds and fertilizers, and services, such as credit. Variations in production have a marked effect on the food security situation among Southern African Development Community (SADC) member states and in the region as a whole. Changes in marketing mechanisms and infrastructure have introduced a temporary period of instability, resulting in swings and shifts in cereal production and supplies in some countries, especially those in which import and export opportunities have arisen. To compound the problem, most of the countries in the region have begun to operate under a Structural Adjustment Programme which, among other things, has resulted in a reduction of government support services and increased unemployment.

The Southern African Development Coordinating Conference (SADCC) was created in April 1980 by nine independent countries in Southern Africa, i.e., Angola, Botswana, Lesotho, Malawi, Mozambique, Swaziland, the United Republic of Tanzania, Zambia and Zimbabwe. Its aims were to promote economic growth through cooperation and to lessen reliance on South Africa. In 1990, the newly-independent Namibia also joined SADCC, which became the SADC in August 1992. South Africa became a member of SADC in August 1994 and Mauritius in August 1995, bringing the membership to 12 countries.

In order to ensure food security throughout the region and to be able to cope with such events as droughts, the SADC Council of Ministers in November 1980 established an initial Programme of Action in the area of food security, which evolved into an SADC Regional Food Security Programme. Zimbabwe agreed to coordinate and implement this programme. The programme secretariat was the Food Security Technical and Administrative Unit (FSTAU) in Harare, officially part of and staffed by the Zimbabwe Ministry of Agriculture. The SADC Regional Food Security Programme had three objectives: to reinforce and facilitate the efficient growth and development of food and agricultural production in member states so that the region could become self-reliant in relation to its basic food needs; to promote the development and harmonization of policies, projects and programmes intended to increase agricultural production, improve household food security, raise rural incomes and facilitate employment generation; and to eliminate the periodic food crises affecting areas or countries in the region and promote intra-regional trade.

The programme included a number of national and regional projects to assist member states in contributing to improved regional, national and household food security. These projects encouraged activities to increase the ability of all people to acquire an adequate diet through improved productivity and increased incomes, especially among the rural population, as a result of the wise use and long-term conservation of the region's natural resource base. However, despite improved food production and food security, the region's potential has not yet been fully exploited.

It was soon realized that the Food Security Programme could only be effective if supported by a timely, accurate and reliable information system able to generate data and information on the food security situation. Because of frequent droughts and other natural calamities, the initial emphasis of food security monitoring was on early warning and crop production assessment. This led to two related projects between 1987 and 1993: GCPS/RAF/223/DEN "Regional Early Warning System (REWS) for Food Security" and GCPS/RAF/232/JPN "Remote Sensing Component of the Early Warning System in SADC Countries", followed by a bridging project, TCP/RAF/2260 "Remote Sensing Component of the Early Warning System in SADC Countries".

The main aim of the REWS project was to establish a food data recording system to monitor the availability of food stocks, progress during the production season and final output. It recommended that a Regional Early Warning Unit (REWU) be established in the FSTAU within the Ministry of Lands, Agriculture and Rural Resettlement (MLARR) in Zimbabwe and that National Early Warning Units (NEWUs) be established or improved in each SADC country. Apart from providing national early warning systems, these would provide inputs to the REWS and act as receiving centres for REWS outputs.

Project GCPS/RAF/223/DEN became operational in 1987. By the end of the project, an REWU had been established in the FSTAU and NEWUs existed, at varying levels of development, in nine SADC countries. Despite this progress, it was felt that more input was required to strengthen some NEWUs and to develop crop assessment methodology. A project encompassing the FSTAU and seven SADC countries (with the other three participating as independently-funded national projects) was therefore prepared and funded. The project ran from 1990 to 1996.

The technology of remote sensing was encompassed first by GCPS/RAF/232/JPN, followed by a bridging TCP project and GCPS/RAF/296/NET "Remote Sensing for Early Warning". The aims of these projects were to establish a remote sensing section at the REWU to serve both national and regional activities, to train counterpart staff in the methodology and to establish an operational protocol for transmitting the analysis and outputs to the NEWUs. The third project expanded the training of the relevant technologies to the appropriate national staff, developed Geographic Information System (GIS) capabilities in the existing NEWUs and integrated remote sensing into the national and regional early warning systems.

However, it was soon recognized that accurate forecasts of adequate food supplies did not ensure that all people had sufficient access to food or the right kinds of food, whether through home production, market purchases or other means. Almost all SADC countries experience food insecurity problems, particularly at household levels in rural areas and among the urban poor and unemployed. Since the nature of insecurity varies from country to country and from one season to another, however, the SADC Food Security Programme was revised in 1987, placing new emphasis on access to food by vulnerable population groups. Increasing attention was given to looking at the food access situation and the nutritional wellbeing of the population.

It was initially planned to attach a household food security and nutrition (HHFS&N) component to the second phase of the Regional Early Warning project (GCPS/RAF/270/DEN), but the donor of the REWS project noted that HHFS&N activities and focus were unique enough to require a separate project. SADC therefore requested FAO to prepare a project document on HHFS&N activities. The Government of the Netherlands indicated its interest but stipulated that the project be oriented towards support to development planning and action programmes and that it concentrate on two to three countries during a pilot phase. It stressed that the assistance be focused on national governments that had demonstrated a commitment towards influencing food access policies and that had shown an interest in support for nutrition.

It was agreed that the project would concentrate on three countries, with Zimbabwe, Swaziland and Angola being mentioned as the primary candidates. Until the end of 1991, the institutional focus of the project was to strengthen NEWUs in the area of food access issues. After a review of the project document by the FSTAU, however, the inputs for improving and developing HHFS&N capabilities were broadened to include other government institutions.

A crucial impetus to introducing HHFS&N issues in the region was a workshop sponsored by the FSTAU, the United Nations Children's Fund (UNICEF), the United States Agency for International Development (USAID), the United Nations Development Fund for Women (UNIFEM) and Cornell University's Food and Nutrition Policy Programme, held in Malawi in April 1992, during which the food security and nutrition monitoring situation and prospects in the SADC countries were discussed. The FSTAU canvassed SADC representatives to solicit interest in HHFS&N and FAO and the Dutch representatives finalized a draft project document. Following the workshop, the Government of the Netherlands agreed to fund a pilot project to develop a regional food security and nutrition information system. Despite the strong national emphasis, the donor also insisted that the project receive significant guidance and overview from a regional component, to be situated in the FSTAU in Harare.

The three countries selected for the pilot project were Angola, Swaziland and Zimbabwe, the latter being chosen in the belief that it would foster the transfer of information to the regional system.

1.2 OUTLINE OF OFFICIAL ARRANGEMENTS

The Project Document was signed by FAO on 26 May 1993 and by the Government of Zimbabwe on behalf of the SADC on 27 May 1993. Initially the project was scheduled to run for two years with a budget of $US 1 880 000 and counterpart contribution in kind. FAO was designated the executing agency and the Government of Zimbabwe (FSTAU at the MLARR), on behalf of SADC, the counterpart agency responsible for project implementation.

As a result of a deterioration in the civil security situation in Angola, FAO and the Government of the Netherlands recommended to SADC that a third country be substituted for Angola, which would remain involved only at a much reduced level. Since negotiations for replacing Angola with Mozambique were prolonged, minimal efforts were made to identify an international candidate for the post of Technical Adviser (TA) in that country. A Selection Committee meeting was conducted in June 1993 for the four internationally recruited positions and the positions of Chief Technical Adviser (CTA) and Socio-economic/Nutrition Adviser or Technical Adviser for Zimbabwe were selected and subsequently approved by the Government in September 1993. The selected candidate for Mozambique was not available, while the Government of Swaziland requested at least two additional candidates to supplement the one candidate presented.

The project became operational in October 1993. Following the two Tripartite Review (TPR) missions in March 1995 and May 1996, the duration of the project was extended by sixteen months to 31 March 1997, with a total budget of $US 1 976 000, subsequently revised to $US 2 208 625. As a result of savings incurred, the donor then agreed to extend project activities, without additional funds, to 30 September 1997. Since the Swaziland component was the last to begin, it was decided to extend the presence of the TA for an additional three months, to the end of June 1997.

1.3 OBJECTIVES OF THE PROJECT

1.3.1 Development objective

The development objective of the project was to improve the design and implementation of specific interventions and action programmes and of national and regional policies to achieve regional food security and improved nutrition for all population groups.

1.3.2 Immediate objectives

- to provide timely nutrition and socio-economic information on vulnerable population groups on a regular basis to decision-makers, providing background information to enable them to design more effective emergency/relief operations, action programmes and local-level interventions, as well as national and sectoral planning according to the needs of the country to enhance and improve nutritional status in Mozambique (originally Angola), Swaziland and Zimbabwe by the end of the project (originally by 1994); and

- to establish an operational framework for the creation of a regional food security and nutrition information system.

2. RESULTS AND CONCLUSIONS

The project successfully identified the institutional linkages and framework of data-providers within the food security and nutrition information system and of those dealing with matters facilitating the use of data. It also identified the decision-makers and their information needs, including that of transforming data and information into a useful form, and provided training to a cadre of analysts who will work with the data and information. The food needs of each country were assessed and a way was prepared for both a regional food security system and the expansion of the pilot project phase.

2.1 INSTITUTIONAL LINKAGES

In all three countries, those institutions and persons whose participation in any food security and nutrition information system would be useful were rapidly identified. A considerable amount of time, however, was required to sensitize potential stakeholders to the importance of actively participating in the system.

A crucial issue was whether the system should be placed in a central or line ministry. Although central ministries have a mandate to oversee the planning activities of line ministries, they are often remote from implementation. On the other hand, line ministries often regard themselves as the owners of food security issues and have difficulty in accepting the involvement of other ministries. Alternatively, they may find it difficult to elicit a response or cooperation from other ministries.

Ownership can be interpreted to mean embracing the vision of the project in terms of the philosophy of inter-sectoral linkages and seeing the project as a network mechanism. In other cases, ownership refers to the process by which the host institution sees the project as reflecting the achievement of its own mandate and activities. Finally, it can describe the controlling of physical assets or tangible benefits, such as training.

Examples of all three interpretations of the term were found in each country. In each country similar lessons were learned regarding the requirements for the establishment of institutional linkages. These included: a readiness for a sustained period of dialogue; the identification of key personalities and points of leverage; an initial willingness to accept ad hoc informal links in the expectation that they would lead to formalization; a flexibility to adapt to the different time-scales of different institutions; and a tendency to centre contact around concrete activities rather than meetings.

A common lesson was that the links established were in a state of constant evolution. This was both positive and necessary, given the weak institutional settings within many of the governments. While experiencing differing levels of success in the three countries, the project was largely successful in building an inter-sectoral HHFS&N information system, albeit in different settings, requiring different solutions.

2.1.1 Mozambique

In Mozambique, an institutional framework with linkages at both national and provincial level was identified and established for the assembly and analysis of HHFS&N information. Food security and nutrition information activities became an integral part of the functions of the Poverty Alleviation Unit (PAU), part of the Department for Population and Social Development (DPSD) in the Ministry of Planning and Finance. Strong links were developed with the National Institute of Statistics, the Nutrition Section of the Ministry of Health (MOH), the Directorate for Agricultural Economics, the Crop Forecasting and Monitoring Unit of the National Directorate of Agriculture, the Food Security Department (FSD) of the Ministry of Industry, Commerce and Tourism, Eduardo Mondlane University, non-governmental organizations (NGOs) and United Nations organizations. The existing links with the Nutrition Section of the Ministry of Health were consolidated with the placement of an Associate Professional Officer (APO) within that unit.

2.1.2 Swaziland

In Swaziland, with the pilot phase of the project firmly established in the Home Economics and Nutrition Section (HEO), strong links were made throughout the Ministry of Agriculture and Cooperatives (MOAC). In particular, the project strengthened institutional linkages within the MOAC through the project Steering Committee (with representation from the Economic Planning and Analysis Section, the NEWU and the Home Economics Department at the Agricultural Campus of the University of Swaziland) and membership on the Early Warning Technical Advisory Committee. Most importantly, it cooperated with the National Early Warning System and the other information-collecting units of the Ministry (e.g., Market Advisory Unit) and was closely linked with policy formulation efforts of the Ministry and with the Ministry's national sensitization efforts on food security in preparation for the 1996 World Food Summit and its continuation in the form of a National Consultative Committee on Food Security.

It strengthened institutional linkages outside the MOAC through active participation in the National Nutrition Council (including the development of Swaziland's Plan of Action for Nutrition), the World Food Day Committee, the National Consultative Committee on Food Security, the National Development Strategy process and through the institution of a collaborative, multi-sectoral task force to determine the shape and direction of the second phase of the project.

2.1.3 Zimbabwe

The institutional linkages in Zimbabwe did not develop as had been anticipated, for a number of reasons. At the time the Project Document was formulated, there were plans to house all food security activities, including the NEWU, in the MLARR. In addition, preparations for the International Conference on Nutrition (ICN) were under way and the two principal ministries with clear food security and nutrition mandates, MLARR and the Ministry of Health and Child Welfare (MOHCW), had begun to collaborate. Before the project became operational, however, a number of events occurred that impeded the success of the Zimbabwe component. The transfer of the NEWU to the MLARR failed to take place. The MLARR became the Ministry of Lands, Agriculture and Water Development (MLAWD) and, in response to the newly-introduced structural adjustment programme, underwent major reorganization. The MLARR and MOHCW gradually began working in isolation. Towards the end of the first year of the project, the MLAWD made it clear that it regarded the project as its own responsibility and showed no interest in expanding ownership to other ministries. In 1995 the MLAWD reorganized once again, separating into the Ministry of Lands and Water Development and the Ministry of Agriculture, with the project remaining with the latter. Finally, the MOHCW gradually lost interest in the project's activities, even though the training provided and the activities planned included MOHCW personnel.

Following the ICN, the MLAWD and the MOHCW considered ways to enhance the profile and influence of HHFSN issues. A MLAWD/MOHCW study was conducted in which two options for the placement of an HHFS&N unit were discussed. In January 1994, the two ministries conducted a one-day seminar during which these options were debated. However, following this seminar, a report and follow-up action were not forthcoming and the issue was not re-examined until late 1995. Although the plan to launch a Nutrition Council was announced in late 1995, the proposal had yet to be considered by Cabinet when the project came to an end in March 1977.

Although the development of a robust HHFS&N information system was frustrated by the situation described above, certain ministries, units and NGOs expressed an interest in working with the project and produced a number of successful outcomes.

2.2 INFORMATION NEEDS

The project devoted considerable time and effort in sensitizing senior-level decision makers and mid-level technicians to the basic concepts of food security and nutrition and the inter-sectoral nature of this issue. In addition, attempts were made to understand what information was needed and what format would be most appropriate for it. This process varied according to the country, in-country support and ownership of the project. Initially it was found that many users of information were not confident in expressing their information needs and were dependent to a certain extent on support from the project. In all three countries it was recognized that sensitization was an ongoing process.

Expectations that considerable amounts of hard, quantifiable nutrition and food security data existed in each country were not confirmed. However, in some cases, the project initiated a process among data collectors that would lead to the availability of useful hard data in the future. The lack of quantifiable information resulted in the development of methodology and training in obtaining softer data, for example, by using Rapid Food Security Assessment, with an emphasis on the participatory approach, and the Save the Children's RiskMap.

2.2.1 Mozambique

In Mozambique, senior-level decision makers and technicians from a number of ministries concerned were sensitized to basic concepts of poverty, food security and nutrition and an ongoing dialogue began concerning the need for, and use of, information for policy formulation and the planning of interventions.

A Poverty, Population and Food Security Data-base was established in the PAU, containing data from sector ministerial information systems and ad hoc surveys. This information will enable the DPSD to monitor the implementation of the Government's annual economic and social plan incorporating plans from the key line ministries. The data-base also provides background information on the food security and nutrition situation from the agriculture, health, commerce, education and social welfare sectors.

Considerable existing data and information is under-utilized. This is partly because recent disruptions have limited the opportunities to gain experience in using information for the formulation of policies and plans, particularly with a long-term view, and partly because of the poor quality and content of existing administrative data sets.

District Food Security and Nutrition Profiles were collated and analysed for 69 of the 131 districts in the country. These provide a unique source of food security and nutrition information for Mozambique. The information is being used at provincial level to design and plan health sector interventions and to sensitize other sectors to the specific food security and nutrition problems which should be addressed. At central level, the information is being aggregated into food economies and a vulnerability mapping data-base is being created, by adapting the RiskMap programme of the Save the Children Fund (SCF-UK), to indicate the characteristics of areas and households subject to food insecurity. This approach to filling data gaps has already yielded useful information. However, the remainder of the country should be completed and food security and nutrition profiles for urban areas compiled.

The data available from the National Household Living Conditions Survey (IAF, 1996, implemented by the National Institute of Statistics) will make it possible to validate and complement administrative data sources and the information provided by the District Food Security and Nutrition Profile.

2.2.2 Swaziland

The Swaziland component devoted considerable time to uniting the data and information needed to assess food security, including the conducting of pertinent ad hoc surveys. One major effort was to digitize various base maps so that food security analysis could be carried out using GIS techniques, including the enumeration areas used by the Central Statistics Office (CSO) in its surveys and the base planning areas used by MOAC's Department of Veterinary and Livestock Services.

The project also estimated the maximum level of disaggregation possible for each set of data in an attempt to determine household-level food security conditions by means of existing secondary data. Because of the dearth of disaggregated data, the Steering Committee asked the HEO to develop and test a survey instrument for its officers that included Participatory Rural Appraisal (PRA) techniques. A timetable was set up for HEO extension officers to continue to assess household food security.

2.2.3 Zimbabwe

During the first two-year period of the project, efforts in Zimbabwe were made to review and utilize data from the CSO. However, this office normally required direct payment in order to access and process the requested data or to modify data that had been collected. As was the case in the other two countries, much of the data were of poor quality and not processed in a timely manner. At the beginning of the project it was hoped that information would be collected by the Research Sub-unit of the Economic Division at the Ministry of Agriculture (MOA), but this unit became progressively weaker and was closed in 1995.

Initially, considerable attention was given to under-five clinic data and, in particular, weight-for-age data. At most clinics, children were weighed periodically, but concern was expressed about the maintenance of the scales and staff training. Individual measurements were summarized at the end of the month as numbers and percentages of "underweight" children and recorded on the T5 form, which also contained the clinic's summary statistics of other public health indicators. After securing available summary statistics, an analysis was conducted jointly with a European Union (EU)-sponsored research project on household food security in Buhera District of Manicaland Province. The interpretation of the results raised strong doubts about the advisability of using this information.

Because of the limited and often poor-quality data, the project devoted increasing effort to conducting rapid food security assessments. This methodology was welcomed by staff members from the MOA, MOHCW, CSO, the Department of Social Welfare (DSW) and the Agricultural Research and Extension Services (AGRITEX). In 1994-95, two training sessions were held in Rapid Food Security Assessment (RFSA) with a participatory focus, each followed by experience in the field. In addition, during the last year of the project, AGRITEX began to expand its food security focus during crop assessment missions to include food and input access issues, relying on the RFSA methodology and training to carry out its analysis. The majority of the funds for the initial activities in this area were provided by the project.

2.3 TRAINING

Training was a major component in the Project Document. A number of regional and national training programmes were conducted, although the needs of the three pilot countries differed considerably. For example, Mozambique had government personnel who had not been exposed to training opportunities and whose level of technical expertise was less advanced than that in the other two countries, although interest was high. Personnel in the other pilot countries had already been exposed to training and were less inclined to accept additional sessions. The initiation of regional training activities was hampered by the fact that the project was not operational in all three pilot countries until April 1995.

During the project period, two regional training activities were conducted. The first was an Introductory Workshop on Household Food Security, held in Maputo from 4 to 8 July 1994, with nine participants from Zimbabwe, Mozambique, Swaziland, Zambia and Lesotho. This workshop was conducted while only the Zimbabwe component was active, although representatives from the other two pilot countries were included. In addition, representatives from Zambia were included, because of the strong national HHFS&N system in that country, and from Lesotho, because of that country's strong interest in the issue. The timing of the workshop took advantage of the fact that the REWU Regional Workshop for NEWU agro-economists was being held from 4 to 15 July. During the second week of the REWU workshop, one day was devoted to an introduction of a household food security perspective in early warning analysis.

The second meeting, a regional workshop on HHFS and Nutrition Data Analysis, Interpretation and Presentation, was held in Harare from 22 to 31 July 1996, with 13 participants from Zimbabwe, Mozambique Swaziland, Zambia and Namibia. The topic of this workshop grew out of a need expressed by the participating countries. Once again Zambia was included because of its experience in the issue. Namibia was just beginning a national, internally-funded project including HHFS&N and therefore requested training support from the regional project. In the case of these two non-pilot countries, travel and daily subsistence allowance were provided by the national projects.

On the occasion of the project's Organizational Meeting in Swaziland in September 1995, the field and counterpart staff decided that training needs assessment (TNA) should be conducted in the pilot countries, with the possible inclusion of Zambia, in order to develop a more appropriate training programme. Action was not immediately taken on this recommendation owing to uncertainties regarding the project's future. However, following the May 1996 Tripartite Review, during which it was suggested that a TNA exercise be conducted, this took place in November and December 1996, involving two regional consultants and a national consultant in each country. The regional consultants planned and prepared the questionnaire and held a briefing meeting with the national consultants and project field staff. Agreed changes were incorporated and the national consultants prepared an interview programme for November in their respective countries. The results were forwarded to the regional consultants, who prepared a report for each country, in addition to a summary report.

In a concerted effort to foster inter-project collaboration, the project's CTA participated in the Annual Organizational Meeting of GCPS/RAF/270/DEN in Dar-es-Salaam, 7-12 March 1994, and Mbabane, 16-21 July 1995. In addition, the project collaborated with the EU-funded Food Security Training Project, also situated in the FSTAU, on a number of occasions. The most concentrated input was to provide some of the training inputs for the course on food security and nutrition data analysis, conducted at Bunda College, Lilongwe, and Zomba in Malawi.

2.3.1 Mozambique

In Mozambique, 33 provincial nutritionists were trained in the theory of food security and nutrition information systems. The training material was revised to constitute a food security and nutrition module forming part of the recognized training programme for medium-level technicians in the Institute of Health Sciences. The nutritionists were given practical training in the methods and techniques for collection, analysis, interpretation and presentation of district food security and nutrition profiles. They are now the only cadre of government staff at provincial level capable of making an integrated analysis of the food security and nutrition situation.

Data analysts from the Ministries of Public Finance, Agriculture and Health were trained in quantitative analysis and interpretation, using preliminary data from the IAF. However, these skills need to be integrated into the work of the relevant ministries. Moreover, the transformation of information into policy options should be reinforced.

2.3.2 Swaziland

The Swaziland component prepared and implemented a four-week course on the training of trainers in PRA techniques for food security. The course was carried out in two two-week modules for 20 trainees from MOAC's Agriculture and HEO extension services and NGOs working in drought relief. A training manual based on this course was published, and a one-week course in basic computer use was prepared and implemented for regional HEO extension officers. The project also funded the designated counterpart's attendance at a three-month course on food security analysis at the Institute for Development Studies at the University of Sussex and carried out a TNA of the government sector on food security issues. The project trained all HEO extension officers in the basics of food security, PRA techniques and the use of assessment tools.

2.3.3 Zimbabwe

In Zimbabwe two RFSA courses were conducted, followed by field training. Reports on both the training and field work experience were prepared. The project also collaborated with an SCF-UK study on the differing coping strategies in the 1992 and 1995 droughts. Information collection was carried out by government and NGO staff trained in RFSA and PRA techniques. The project also collaborated with the International Red Cross in training selected NGO employees from the SADC region to assess food security problems and devise possible remedial strategies at the local level.

In discussions with government personnel dealing with nutrition-related issues, it was seen that skills in rudimentary computer tasks were deficient and that the computer equipment available often did not allow the use of state-of-the-art software. It was decided to hold a series of short (half-day) training sessions on simple DOS commands and software exploitation, with particular emphasis on table design and data presentation.

2.4 FOOD NEEDS ASSESSMENT

In all three countries, national energy requirements were calculated. In addition, during both the regional workshops mentioned above, instruction in the use of the appropriate programme was given, as well as an explanation of the principles behind energy requirements. The Swaziland component contracted the University of Swaziland to calculate energy requirements and to collect local data pertinent to energy requirements in three agro-ecological zones.

2.5 FACILITATING A REGIONAL FOOD SECURITY SYSTEM

During the project, the CTA assisted non-pilot project countries in setting up an HHFS&N system or maintaining and promoting existing systems.

At the SADC Council of Ministers meeting in Pretoria, on 14 August 1995, the FSTAU was mandated to expand the Regional Food Security and Nutrition Information System project as soon as possible to all SADC member countries. With this directive, the highest reviewing body of SADC endorsed the second phase of a Regional Food Security and Nutrition Information System.

Following the acceptance in January 1996 of the recommendations of the April 1995 TPR Mission, it was decided to build a regional food security and nutrition information system by assisting the member countries of SADC to prepare their own project documents. The Sector Coordinator of the FSTAU insisted on the full involvement of in-country personnel in this task. In December 1996, a workshop was held in Harare, attended by representatives from the different countries to exchange experiences in developing a Phase II project.

Since Angola had originally been included in Phase I, a strong commitment remained that it should be involved in the second phase. Independently of the FSTAU, FAO prepared a project document titled "Assistance au D�partment de la S�curit� Alimentaire (Gabinete de Seguran�a Alimentar GSA)", GCPS/ANG/024/NET. In August 1997, the CTA explored ways to introduce an HHFS&N perspective, with particular attention being given to the methodologies developed in Mozambique.

Botswana expressed a strong interest in developing an HHFS&N information system. In early 1996, discussions were initiated with the Permanent Secretary in the Ministry of Agriculture. A task force was formed to discuss the relevant issues and to prepare a draft project document. The CTA was asked to attend a workshop on 23-24 May to further develop the project document, which was then given to the CTA to incorporate additional points made at the task force meetings.

Lesotho was one of the earliest proponents for joining the project. The CTA visited Maseru in October 1994 and prepared a draft project document for TCP funding designed to provide a bridge for subsequent UNDP funding. However, TCP funding was not secured and interest waned. In 1996, the Government made it clear that it did not want additional studies, but assistance in formulating project documents that assisted in specific interventions. The CTA prepared a short project description which was shared with the Government. The Food Monitoring Unit prepared a preliminary first draft, requiring substantial work.

The CTA visited Malawi in early April 1996. During the visit, the Government appointed a task force and assigned personnel to prepare the document. It was agreed that a workshop would be held, after which the project document would be finalized. The workshop was held in Salima, on 25-26 April, and was attended by a representative from the World Food Programme (WFP), the USAID key adviser to MOA, UNICEF, MOA, UNDP, MOH and Bunda College. A draft project document was circulated to the Government and then sent to FSTAU. The CTA visited Lilongwe and held a short meeting with the potential stakeholders, at which it was agreed that the two principal ministries (the MOA and the Ministry of Economic Planning and Development) would again try to reach an agreement on the institutional arrangements.

After joining SADC in 1995, Mauritius expressed an interest in being part of the HHFS&N information project. The food security issues concerning Mauritius are better food control and the effects on health concerns of an "affluent" society diet, i.e., heart disease, hypertension, diabetes, obesity and high blood cholesterol. Any monitoring or information system would thus be substantially different from those in other SADC countries. No attempt was made to develop a project document.

In Mozambique the project's TA facilitated the preparation of the project document. Extensive discussions with potential partners were held and there was general support for a second phase. The project established particularly close links with the MOH's nutrition unit and much of Phase II will be targeted to decentralizing the monitoring process through the provincial health workers. Inputs and comments from the CTA and FAO Headquarters were provided and the completed document was presented to the Netherlands Embassy in Maputo for funding. Funding was secured and the new project became operational in June 1997.

Between August 1994 and July 1995, Norway funded an FAO-operated project in Namibia titled "Policy and Programming Assistance for a National Food Security and Nutrition Action Plan". One output was a project proposal titled "Support to Food Security and Nutrition Development in Namibia". This proposal was approved at the highest levels in the Government, internal funding of $US 1.1 million was committed and a Food Security and Nutrition Unit in the Ministry of Agriculture was established and staffed. The project became operational in June 1997. The Government is also soliciting external donors for additional activities.

In South Africa, where the Ministry of Agriculture is shifting its emphasis from concerns with commercial farms to those of "communal farmers", initial contact was made with the Land and Agriculture Policy Centre in Johannesburg and the Reconstruction and Development Programme in the Office of the President. After consultation with the Sector Coordinator, it was decided that attention would not be given to HHFS in the immediate future.

In Swaziland the TA facilitated the preparation of a Phase II project document. Five task-force meetings were held, an institutional framework was agreed on and a project document prepared. A final project document has been submitted to the EU for funding consideration.

Following repeated requests for assistance in preparing a Phase II project document, the CTA visited Dar-es-Salaam to assist the Government of Tanzania. After a workshop, discussions were held with government officials and a draft project document prepared which concentrated on a decentralized system, since it was this aspect in which both potential donors and the Government expressed most interest. The draft was then reviewed at FAO Headquarters and within the Government, before being shared with an Irish funding mission. The EU and the Dutch Embassy also expressed an interest in the focus.

The FHANIS (Food, Health and Nutrition Information System) in Zambia is already a Phase II project, following a Phase I TCP project (TCP/ZAM/2353) from February 1993 to July 1994. The Phase II project is currently two-thirds funded and it is planned to seek a third donor in addition to UNICEF and IFAD. Following FHANIS's move from the CSO to the National Commission for Development Planning (NCDP), the Commission created a new unit titled Human Development to house FHANIS and food security and poverty concerns. Since NCDP staff were unfamiliar with food security issues, it was felt that a project document could be designed around strengthening the staff of this unit. While this was being prepared, the NCDP was split up and various functions assigned to different ministries. The Human Development Unit and FHANIS were placed in the augmented Ministry of Finance and Economic Development. The Netherlands, the EU, Ireland, the World Bank and USAID have expressed interest in providing support.

In Zimbabwe the institutional framework problems that impeded the implementation of Phase I also affected the design of an effective Phase II project. The Sector Coordinator of FSTAU introduced the FSTAU as a partner and objective broker in the discussion concerning a Phase II document. This arrangement was accepted by the potential stakeholders and led to an initial meeting. However, agreement was not reached, frustrating attempts by the FSTAU to produce a draft Project Document. The task force for a Food and Nutrition Policy recommended that a Nutrition Commission be housed in the National Economic Planning Commission, but the MOA wished to retain full responsibility, while the FSTAU favoured either the National Economic Planning Commission or the Ministry of Finance. The Sector Coordinator insisted that the resolution of the institutional framework and early drafts of the project document be produced by nationals.

Development of a sustainable HHFS&N information system at the regional level was hampered by the lack of a regional counterpart and questions were raised regarding the need for a regional entity at the FSTAU. During the final year and a half of the project, major donors urged the FSTAU to develop a total programme in early warning and food security, identifying the various components and uniting them into a total funding programme package. During late 1996 and early 1997, the EU provided funds for carrying out a total FSTAU programme analysis. Although HHFS&N played a primary role in the suggested strategy for FSTAU, delays occurred in putting together a unified proposal.

3. RECOMMENDATIONS

3.1 GENERAL RECOMMENDATIONS

The most crucial condition for the ultimate success of an inter-sectoral project is the establishment and maintenance of the correct institutional linkages and framework. If this is unsuccessful, it is unlikely that the remaining objectives will be achieved. The process begins with the project document, which should be prepared with the involvement and commitment of a true inter-sectoral representation from the country. This may be difficult if some government staff members are unfamiliar with the principles of project design or do not wish to develop a project whose direct benefits they may not understand. Consideration should therefore be given for the project to operate initially in a pre-implementation phase, the main aim of which is to establish the credibility of the activity and those associated with it. During this phase, the institutional arrangements should be worked out and the potential stakeholders sensitized to the project's objectives, outputs and benefits, as well as to the role that each can play. Country personnel, drawing upon an inter-sectoral team, should be full participants in the preparation of the project document. Furthermore, sufficient flexibility should be allowed in project activities to make even major adjustments in the institutional linkages and framework after the project has become operational.

Although the project may be inter-sectoral, it is often housed in one unit. This can lead to that unit becoming over-possessive of the project, thus discouraging an inter-sectoral spirit. In addition, units not directly involved may prefer to remain so, fearing an increase in the work load or conflict with other units. Tangible inputs to, and benefits from, the project should therefore be distributed beyond the unit which houses the project, with full consideration being given to the project's overall needs and outputs.

When assigning staff to project activities, insufficient consideration is often given to current work loads or commitment incentives. It should be accepted that many government personnel are underpaid, overworked and operating in an environment of retrenchment. Project documents do not always reflect this situation and can portray a false impression of full commitment by the government to the project. It is recommended that a system of established incentives and rewards to the staff working with the project and their supervisors be negotiated during project preparation in order to ensure project commitment. One option would be to link project activities and the skills learned and applied during the project with promotion and government pay enhancements.

The stipulation to assign a government counterpart to the project is often routinely included in a project document. Most often the counterpart responsibility falls to one person. Although there is an assumption that this person will work full-time on project activities and continue to develop the work and expertise acquired once the project period has ended, this is not always the case. In addition, even in those instances where the counterpart does fulfil these conditions, he or she may be transferred or seek different employment at a future time. When this occurs, there is no one with the responsibility and knowledge required to assume the role. It is therefore recommended that the concept of a team of counterparts, rather than that of a single counterpart, be developed, since this will allow the spreading of expertise, knowledge and training over a broader base.

All HHFS&N activities must encompass the identification of decision-makers and their information needs. Any future project design should therefore incorporate an ongoing process for identifying decision-makers, assisting them in identifying their own information needs and bringing them into the HHFS&N information system.

Although it is acknowledged that any regional system must be built on strong national systems, these national systems, in turn, require contact with each other and a level of coordination from the regional level. It is therefore recommended that future HHFS&N activities contain a viable regional component, housed in the FSTAU and with its sustainability ensured by the SADC.

In order to play its coordinating and assistance role adequately, the FSTAU must be a viable and strong body. Regional staff should be well-versed in technical skills with a regional perspective, while their positions should be secure and well-remunerated. Any future assistance should include means for strengthening the FSTAU in those areas relevant to HHFS&N.

Since training is critical in the development of any HHFS&N system, all training needs must be carefully worked out with the stakeholders, in close collaboration with the recipients and supervisors, and should be owned to such an extent that the owners are willing to pay, in full or in part, for their training. The training procedures themselves should be developed by a qualified trainer and be closely related to the developed work programme. It would be beneficial if all training received could be formally incorporated into career development. This would require high standards, both of training and of the outputs expected from participants.

Discussion with donors revealed that increasing importance is being given to decentralized HHFS&N systems, since it is felt that such systems may be more sustainable and have a more immediate impact. Future HHFS&N systems should therefore contain a decentralized component or at least develop in that direction.

3.2 MOZAMBIQUE

In order to ensure that food security and nutrition information is acted on, it is recommended that the Government establish a formal mechanism enabling information to be discussed in an inter-sectoral forum and channelled to the appropriate decision-makers. This would help to ensure that policies and programmes are complementary across different sectors and time frames.

In view of the Government's decentralization policy, it is recommended that the collection, analysis, interpretation and use of data be strengthened at provincial and district level.

In order to guarantee the sustainability of food security and nutrition information systems it is recommended that key knowledge and skills related to the subject be integrated into the curricula of pre-service training institutions. Bibliographic and reference material to enable continued in-service training should also be provided.

Many food security and nutrition problems faced by Mozambique are common to neighbouring countries. It is recommended that technical and policy issues be discussed at the regional SADC level in order to exchange experiences and information and identify complementary solutions.

3.3 SWAZILAND

A formal mechanism for the discussion of information and the channelling of that information to appropriate decision-makers in an inter-sectoral forum is recommended. It is also recommended that formal consultative measures and coordinating mechanisms be established to ensure that agreed-on data is collected and delivered in a timely manner.

Continuing efforts should be made to strengthen the culture of inter-sectoral collaboration. This should include the formal fostering of the development of a National Plan of Action on Food Security. The strengthening of data collection, analysis, interpretation and use at the decentralized regional level is also recommended.

3.4 ZIMBABWE

Only limited HHFS&N activity can be carried out until an institutional framework is established and accepted by the Government. Until such a framework is established, any activities pertaining to Zimbabwe should be carried out under the auspices of the FSTAU.

The collection, handling and analysis of HHFS&N information and data require extensive enhancement throughout the Government. It is recommended that training needs be developed in the context of the work programme of the staff and include the active participation of the supervisors.

Appendix 1

Appendix 2

The aim of the National Demonstration Centre (NDC), organized by the Statistics Division, FAO Headquarters, is to train officials from English-speaking countries of Eastern and Southern Africa in the conceptual bases and practical solutions to problems associated with the use of food supply and consumption indicators to assess and monitor the food situation. Twenty-seven participants (15 from Mauritius) from the Ministries of Agriculture or the Central Statistical Offices of Botswana, Kenya, Lesotho, Mauritius, Seychelles, the Sudan, Swaziland, Tanzania, Uganda, Zambia and Zimbabwe attended. When it was realized that all the representatives from Zimbabwe came from the MOHCW, project funds were made available to include the new project counterpart from the MOA.

IDS offers an annual twelve-week course in Food Security in Africa to help food planners and practitioners in African countries analyse the causes of food insecurity in Africa, take a critical look at policies and interventions to alleviate food insecurity and prepare for attempts to tackle food insecurity in the future. A wide range of topics is covered, including food production and employment issues, food marketing issues and famine and emergencies.

The overall aim of the programme was to improve capacities in SADC member states for the analysis and interpretation of data on food security and nutrition. The direct objective was to provide participants with the main skills required to analyse and interpret household survey and other data on food security and nutrition. A subsidiary objective was to develop a set of training materials which could be used in national and sub-national training programmes to strengthen analytical capacity. Both the CTA and the National Project Manager from Zambia contributed by organizing and running some of the sessions. A final version of the training materials has been prepared by the FSTP for future training.

The Multi-media Communication Campaigns for Development enabled participants to prepare communication campaigns for implementation, by building communication skills and putting them into practice. This means that development projects which enrol their staff as a team gain professionally-designed pilot communication campaigns to promote their objectives with rural communities and that all participants acquire skills that can be used for future activities.

One Swaziland-specific recommendation made by the project-wide Training Needs Assessment completed in November 1996 was that Swaziland send a team to attend the FAO/SADC Communications Project's April-June 1997 workshop on Multi-Media Communication Campaigns. An important need in Swaziland was the sensitization of decision-makers and yet the TNA report questioned the effectiveness of past efforts in this area. It was anticipated that the skills learned through this workshop would improve the ability of those working in household food security and nutrition to carry out this task on their own. Thus, the workshop not only provided team members with the intended skills but also provided training and experience in writing proposals and seeking donor funding.

Three officers attended the three-part workshop (three weeks of training; four weeks of local research, including instrument development, data collection and report writing; three weeks of further training aimed at preparation of a multi-media communication campaign). The officers included the project counterpart, the information officer from within the Home Economics and Nutrition Extension Section, and an extension officer from the Ministry's Crop Storage Section, which has nationwide coverage. The team successfully completed workshop-driven (but project-funded) field research in one area, chosen for its remoteness -the theme chosen by the team was that of how well the extension message on crop and food storage was reaching remote areas.

Based on their findings, and with the assistance of workshop facilitators, the team developed a communication strategy and produced two funding proposals: one for funding the campaign itself and the other, titled "The Nzongomane Coalition," for mobilizing the expected campaign participants.

Appendix 3

Appendix 4

Report of workshop on household food security concerns in Maputo. W. Bender. July 1994.

Anthropometry tutorial and references information computer tutorial. April 1996.

Training materials from regional workshop on HHFS and nutrition data analysis, interpretation and presentation. S. Devereux. Harare, 22-31 July 1996.

Report on training needs assessment for food security and nutrition information systems in Mozambique, Swaziland, Zimbabwe and Zambia. K. Selvester and L. Fidalgo. December 1996.

Compendium of project experience in Rapid Food Security Assessment and Participatory Rural Appraisal. K. Selvester. September 1997.

Rapid Food Security Assessment: Proceedings of a training workshop. Harare, 1994.

Rapid Food Security Assessment exercise in Matobo District, Matebeleland Province: Methodology, results and recommendations. 1994.

Rapid Food Security Assessment in Zvishavane District, Midlands Province, Zimbabwe: Methodology, results and recommendations. October 1995.

Conducting Rapid Food Security Assessments: Guidelines for trainers and researchers. 1995.

Methodology for the compilation of District Food Economy Profiles: Guidelines for field work, analysis, interpretation and presentation. Portuguese. Parts I and II. June 1996.

Training module: "Integrated Food Security and Nutrition Information Systems". Portuguese. April 1997. (Draft: final version in preparation.)

Basic instruction manual for DOS, WordPerfect 6.1, Epinfo6 and Lotus 123 for Windows. Portuguese. August 1996.

Training Modules: "Data Analysis and Interpretation for the Formulation of Food Security and Nutrition Policies":

Module 1: General introduction to poverty, household food security and nutrition concepts. November 1996.

a. General concepts on poverty for decision-makers and data analysts. Portuguese. S. Cassamo.

d. Not by information alone....information needs for policy formulation. Portuguese and English. M. McEwan.

Module 4: Interpretation and policy formulation: presentation of results. Portuguese. May 1997.

Preliminary assessment of the use of the SCF-UK RiskMap approach for mapping vulnerability in Mozambique: a case study from Cabo Delgado. English. May 1996.

Regional Workshop on HFS and nutrition data analysis, interpretation and presentation: evaluation reports. I. Macu�cua and P. Mavume. Portuguese. August 1996.

Conceptual and operational framework for the Household Food Security and Nutrition Information system in Mozambique. English. May 1997.

Information-users' needs assessment in Mozambique: An evolving dialogue. English. May 1997.

Inventory of food security and nutrition data sources in Mozambique. English and Portuguese. May 1997.

A rural poverty profile for Mozambique. (UAP). English and Portuguese. April 1996.

District food economy profile summaries. Portuguese. (25 finalized, 35 in preparation.)

Provincial summary of district profiles for Cabo Delgado Province. Portuguese. August 1996.

Characteristics of rural household food security and nutrition in Mozambique. English. April 1996.

District profiles on food security and nutrition in Mozambique. English. S. Meershoek. Rome, 9-10 December 1996.

Proceedings of Conference on Food Economies in Mozambique: Food security and nutrition vulnerability mapping. Portuguese (with English commentary). Maputo, 6-7 February 1997.

- food security and nutrition situation in Mariend Ngouabi and `7 de Setembro' villages, Boane District: A comparative analysis. Portuguese. I. Macu�cua. January 1997.

- analysis of food security and nutrition in Homo�ne, Inhassoro districts and Bazaruto Island, within the Participatory Poverty Analysis in Mozambique. Portuguese. P.J. Mavume. January 1997.

- food security and nutrition indicators in Mozambique. Portuguese. I. Macu�cua. February 1997.

- evaluation of the sensitization and training seminar on the analysis of household food security and nutrition data. Portuguese. I. Macu�cua and P. Mavume. 26-27 November 1996.

Report on effect of floods in Central Mozambique. English. S. Meershoek. February 1997.

Thematic maps on food security and nutrition for Mozambique. English and Portuguese. May 1997.

Preliminary literature review for the preparation of provincial household food security profiles. Portuguese. J. Manja. August 1995.

Report on in-service training for software packages. English. S. Subramaniam. July 1996.

Literature review: Food and nutrition situation in urban areas in Mozambique. Portuguese. K. Selvester and L. Fidalgo. December 1996.

Training Needs Assessment report for Mozambique. Portuguese. K. Selvester and L. Fidalgo. January 1997.

Project Document for a Phase II project in Swaziland "Household Food Security and Nutrition Information System".

Participatory Rural Appraisal Techniques for studying household food security. K.J.B. Keregero. 1996.

Report of the participatory rural communication appraisal and baseline study, Nzongomane Community, Shiselweni Region. 5-17 May 1997.

Food Security Sector of the Agriculture, Land and Rural Development Sector's submission to the Swaziland National Development Strategy. 1996.

Percent of years that maize yield is more than 80% of the best possible yield. July 1997.

Percent of years that maize yield is less than 50% of the best possible yield. July 1997.

Rainstations and agro-ecological zones (per land-use planning section). June 1997.

MOAC/Department of Veterinary and Livestock Services (DVLS) diptanks and livestock planning areas. March 1997.

Blow-up of tenure types per CSO, showing census enumeration area boundaries. January 1997.

Administrative districts, agro-ecological regions (CSO) and tenure type. May 1996.

DEVELOPMENT OF A REGIONAL FOOD SECURITY AND NUTRITION INFORMATION SYSTEM

Table of Contents

LIST OF ABBREVIATIONS

1. INTRODUCTION

1.1 PROJECT BACKGROUND

1.2 OUTLINE OF OFFICIAL ARRANGEMENTS

1.3 OBJECTIVES OF THE PROJECT

1.3.1 Development objective

1.3.2 Immediate objectives

2. RESULTS AND CONCLUSIONS

2.1 INSTITUTIONAL LINKAGES

2.1.1 Mozambique

2.1.2 Swaziland

2.1.3 Zimbabwe

2.2 INFORMATION NEEDS

2.2.1 Mozambique

2.2.2 Swaziland

2.2.3 Zimbabwe

2.3 TRAINING

2.3.1 Mozambique

2.3.2 Swaziland

2.3.3 Zimbabwe

2.4 FOOD NEEDS ASSESSMENT

2.5 FACILITATING A REGIONAL FOOD SECURITY SYSTEM

3. RECOMMENDATIONS

3.1 GENERAL RECOMMENDATIONS

3.2 MOZAMBIQUE

3.3 SWAZILAND

3.4 ZIMBABWE

Appendix 1

Appendix 2

Appendix 3

Appendix 4