Case studies

IMPROVING EMERGENCY SEED RELIEF IN KENYA: A COMPARATIVE ANALYSIS OF DIRECT SEED DISTRIBUTION AND SEED VOUCHERS AND FAIRS- Paul A Omanga and Paul Rossiter

Paul A Omanga
Catholic Relief Services, Kenya

Paul Rossiter
FAO Kenya

Introduction

Seed aid is a relatively new response in the agricultural emergency and development repertoire in Kenya. For the last ten years, seed emergency operations carried out in Kenya have mainly involved the transfer of seeds from commercial companies (formal seed sector) to beneficiaries (resource-poor farmers in drought- and flood-stricken areas of the country). This approach has transformed the beneficiaries (resource-poor farmers) into receivers, instead of enhancing their capacity to produce and access seed themselves.

The seed aid operations involved many institutions that were often uncoordinated and lacked expertise in local and national seed and crop systems. Consequently, this approach has remained inefficient in addressing the real constraints affecting the seed systems of resource-poor farmers.

The majority of recipients of seed aid are resource-poor farmers living in marginal, drought-prone and remote areas, inadequately served by both the formal seed sector (except in cases of emergency) and formal agricultural extension services. However, their seed insecurity can be alleviated and even overcome if seed aid operations have clearly defined, short-, medium- and long-term objectives, focusing on the chronic seed insecurity in the target communities. Such objectives should include access to agricultural extension services and specifically should aim to:

• focus on capacity building, including establishing institutions and forming partnerships with other relevant institutions, instead of relying on seed handouts;

• broaden farmers' crop genetic resources by exposing them to more appropriate crop germplasm from relevant sources, such as from other farmers, national and international agricultural research centres and appropriate commercial seed companies;

• strengthen communities' capacity in terms of the production, storage and distribution of quality seed, using the existing seed dissemination mechanisms;

• respond to severe seed shortages in a more participatory and decentralized manner, using local capacity, and with external efforts being designed to support rather than supplant.

The purpose of this paper is to (1) discuss community seed based efforts in the arid and semi-arid lands (ASALs) of eastern Kenya; (2) discuss the current efforts to improve seed relief-work in the ASALs; and (3) compare and analyse conventional seed procurement and distribution (CSPD)^[7] and seed vouchers and fairs (SVF) approaches in eastern Kenya in relation to lessons learned from past experience.

Background

Of Kenya's total land area of 583 000 km², ASALs, characterized by low and unreliable rainfall, account for approximately 80 percent. The ASALs are spread over the seven provinces, cover 33 districts and support over 20 percent of the total human population, 50 percent of livestock herds and 5 percent of agriculture output. All 13 districts in eastern Kenya are ASALs and are categorized as agro-ecological zones 3 to 7 (Table 1), based on the classification of Braun (1982).

TABLE 1
Characterization of the districts in eastern Kenya based on cropping and livestock activities

Source: Government of Kenya, 1997-2001.

The livelihood of the inhabitants of Eastern Province, about 3.8 million people in total, is mainly small-scale subsistence-based agriculture. Crops and livestock production are both important parts of the farming system and form the main sources of food and income of over 90 percent of the population. Farm size varies between 2 and 7 hectares (ha) per household, the size increasing as one moves to drier zones. The cropped family land ranges from 30 to 50 percent, again depending on the zones, leaving the rest of land for livestock grazing.

In the subsistence agriculture common in the ASALs, farmers produce a broad range of crops and varieties to meet their basic needs and avoid total crop failure (that may result from the many existing biotic and biotic stresses). Major crops grown include cereals (maize, sorghum and millet) and grain legumes (beans, pigeonpea, cowpea and green gram). Cotton, cassava, sweet potatoes, sunflower, Dolichos lablab, castor, gourds and chickpeas are also grown as part of the common mixed farming system.

Drought caused by low and poorly distributed rainfall is the major cause of low agricultural production in the region. For effective dryland crop farming, maximum utilization of available moisture is a necessity for success even in the best weather conditions. Thus, accessibility of seed at the right time is essential for ensuring good crop production and food security. However, a proportion of farmers frequently experience seed shortage following extended droughts, when seed stocks are used for food because crops fail. The government and NGOs often respond by distributing relief seed, which in some cases can be exotic and unsuitable for local conditions. In cases where locally adapted crop varieties are distributed, these are restricted to the limited varieties available from seed companies. This practice leads to loss of locally adapted crop varieties and a continued reliance on both food and seed relief.

Seed delivery systems in Kenya
Seed delivery systems in Kenya can be grouped into formal and informal systems.

Formal seed system

This refers to the production, processing and packaging, labelling and marketing of certified seed by registered producers - normally private or public seed companies with outlets in many parts of the country. There are about 38 registered seed companies in Kenya, the leading ones being the Kenya Seed Company, East African Seed Company, Western Grain and Seed Company and Faida Seeds. Most of these companies produce cereal seed, especially maize, wheat, barley, sorghum and legumes, especially bean and vegetable seeds. Except for maize, certified seed accounts for less than 5 percent of the seed sown in ASALs. Very little certified seed of open-pollinated crops, such as pigeonpea, cowpea, sorghum, millet and green gram (usually grown by resource-poor farmers who live mainly in ASALs), is produced by private seed companies. They also do not produce seed of vegetatively propagated crops. Private companies are profit driven and consider the seed of ASAL crops expensive to produce and market and the demand unreliable.

Informal system

This refers to the production, processing, marketing and/or distribution of seed by unregistered seed producers. Seed produced is variable in quality and not supported by a certification scheme. Production and marketing are often localized and based on low input technology. Key players in this system include NGOs, farmers, farmer groups and community-based organizations (CBOs). It produces local landraces, improved open-pollinated varieties and a blend of the two. It includes farmer-selected and saved seeds, local market grains and farmer/NGO/CBO-managed seed production programmes.

In Kenya, this system has been producing seed of open-pollinated varieties of cereals, grain legumes and also of vegetatively propagated crops such as sweet potatoes and cassava. The informal system accounts for over 90 percent of the seed sown in ASALs. Because it is based on rain-fed cropping systems, it is highly vulnerable to drought stress, occasioning severe shortages of seeds. Producers have limited access to breeders and basic seed of improved varieties. However, it has potential for sustainability, partly because it is derived from traditional systems and has limited demand for external inputs.

Farmers' seed sources

For dryland crops such as pigeonpea, cowpea, mung bean, common beans, sorghum, millets, open-pollinated maize, and tuber and root crops, farmers in arid and semiarid areas obtain seeds from local sources, especially own-saved seeds, social networks (relatives, and other neighbouring farmers), and local traders (shopkeepers and open-air market grain traders). The majority of farmers in marginal areas rely on seed saved from their own harvests and continue recycling seed as long as the harvest is "adequate" and they are able to keep some for subsequent seasons. Local traders play a critical role in rural communities by purchasing grain at harvest, storing it and selling it back later to the same farmers, either as food or as seed at planting time. These seed institutions are critical to the livelihoods of the marginalized and poor households in the supply of food and seed grains.

Community-based seed activities

In order to improve farmers' access to seed of locally adapted and improved crop varieties, governmental organizations and NGOs have been carrying out community-based seed interventions in arid and semi-arid areas. These interventions include seed and cultural fairs, community-based seed bulking, community-based grain banks and seed recovery and banking.

Seed and cultural fairs

Crop diversity is increasingly recognized as crucial for improving food production and food security among marginal farming communities in semi-arid areas. Management and preservation of seed, based on farmer skills and community participation, could help conserve crop diversity and improve food security among farming communities in drought-prone areas. Although farmers tend to conserve crop varieties that give them better yields, activities that reinforce appreciation of their existing crop diversity need to be supported. One such activity is local community seed and cultural fairs.

Seed and cultural fairs offer a venue for displaying crops commonly grown in the region. During the cultural fairs, traditional foods and tools used in farming within the region are displayed. These include land preparation tools, weeding tools and traditional seed storage structures such as baskets, pots and gourds.

The objective of seed and cultural fairs is to help farmers appreciate their large crop diversity by encouraging them to exhibit planting materials. Farmers have the opportunity to access crop varieties that they need and also to exchange and share information and experiences on farming under their local conditions. The seed and cultural fairs also provide a suitable forum for farmers to share information and exchange seeds beyond their locations, thus giving them access to a wider choice of varieties and maintaining a high level of crop diversity.

Community-based seed bulking

Community-based seed production seeks to involve small-scale individual farmers, farmer groups, NGOs and governmental organizations in forming small but effective seed multiplication units with the objective of supplying quality seeds for farmers' own use and for sale to other farmers. Activities include the selection of farmers or farmer groups to be involved in seed bulking and training on seed multiplication techniques and marketing. The Kenya Agricultural Research Institute (KARI) Katumani Seed Unit, based at the National Dryland Farming Research Centre, provides both the required training and foundation seed to the community-based seed farmers.

Community-based grain banks

This is a system where the grains produced locally are stored and distributed to participating farmers as seed at planting time. Community grain banks are managed by farmers, with supervision from CBOs and local NGOs. If run properly, they alleviate shortages of seed and ensure timeliness of seed supply to rural farmers. They can also act as a safety bank for seed, especially in times of drought. The success and sustainability of these grain banks is of paramount importance for local seed security.

Seed recovery and banking programme

In this system, farmers are facilitated to prepare a list of crops and varieties they require, with emphasis on reliable dryland food crops. The crops are procured from local seed merchants (local seed markets) and kept in a central store (seed bank) at village level. These are then distributed to farmers at planting time. After harvest, farmers return twice the amount of grain or seed given.

Village committees manage the seed recovery and banking with the assistance of location extension staff from the Ministry of Agriculture and local CBOs and NGOs. These CBOs and NGOs usually give the communities initial funds for seed procurement. Each seed bank is run at village level. The village committees monitor the crop in the field and establish and manage the seed banks. Sublocational committees monitor the village committees and provide a forum for sharing experiences.

Emergency seed distribution in Kenya: an overview

Emergency seed is a relatively new development in Kenya. It started in 1992 as a collaborative effort between the government, NGOs, CBOs, farmers and other development agencies to supply seeds to communities faced with acute seed shortage following drought-related stress. In the last decade, considerable quantities of maize, bean, cowpea, sorghum, pigeonpea and green gram have been acquired from seed companies and local markets for distribution. Although intended to be a limited, onetime intervention, emergency seed distribution has become a regular source of seeds for ASAL communities as droughts are repeated and frequent.

TABLE 2
Amount of seed, value and numbers of households receiving seed aid in Kenya 1992-2001

*Seed distributed through seed vouchers and fairs approach
Source: Otado and Ingosi, 2002.

Emergency seed distribution in Kenya from 1992 to 2001

The Government of Kenya through the Emergency Drought Recovery Programme has distributed relatively large quantities of emergency seed to farmers in ASALs since 1992, with a distribution nearly every year. Several NGOs have complemented the government efforts in seed distribution in specific locations and seasons. The amount and value of seeds of various crop varieties distributed between 1992 and 2001 and the number of beneficiaries (households) are presented in Table 2.

In 2000-2001, two approaches were used to distribute emergency seed to the drought-affected communities in eastern Kenya. These included the conventional seed procurement and distribution (CSPD) and seed vouchers and fairs (SVF) approaches.

Conventional seed procurement and distribution approach

In this approach, the Government and NGOs request seed bids from locally registered seed companies. The Ministry of Agriculture or the implementing agency, based on the ability of the seed companies to supply the types of crop, assesses the bids, variety and amounts required, packaging and the unit cost. Successful bidders transport the seed to the affected district areas, where it is received by the implementing agency for storage, awaiting delivery to the divisions and finally to locations where it is distributed to the beneficiaries. The Divisional Officers and local chiefs distribute the seed to the beneficiaries at the Divisional or locational levels.

Where the NGOs are involved, the seed from the seed companies is delivered to the NGOs' local offices for subsequent distribution to the farmers. NGO staff supervise the distribution, often in collaboration with Government of Kenya staff.

Seed vouchers and fairs approach

Seed fairs are special markets organized for the local farmers, grain traders, seed stockists and seed companies to distribute grains as "seed" or certified seed to seed-needy households. Seeds from local, informal and formal sources are marketed at a special market to be exchanged for either cash, or by barter for vouchers.

The affected communities identify seed-needy households based on their own set criteria. The households are then issued with seed vouchers of a predetermined monetary value. Farmers and local traders with surplus grain to be sold as seed are sensitized to bringing a surplus to a selected seed fair site. Voucher holders then exchange their vouchers for seed of crop varieties and quantities of their choice, depending on the monetary value of seed vouchers. On completion of the seed fairs, seed vendors redeem the vouchers for cash.

Implementation of seed distribution programmes: comparing SVF and CSPD

Area targeting

Targeting for the CSPD approach is mainly used at the regional level based on occurrence of drought. Little effort is made to identify the most affected divisions and locations.

TABLE 3
Number of households requiring seed in eastern Kenya in May 2001 and those given seed through SVF in various districts

Source: CRS/FAO, 2001-2002.

In the SVF approach, the districts are chosen based on severity of food insecurity and occurrence of drought. In each district, specific divisions and locations are selected based on rainfall and crop performance during the previous two seasons and the perceived seed availability.

Beneficiary targeting

In the CSPD approach, targeting beneficiaries depends on the agency distributing the seed. Some NGOs follow criteria set by officials within their grassroots networks. The government in most cases offers blanket distribution mainly to satisfy political interests, though, theoretically, frontline agricultural extension staff are supposed to target seed-needy households.

In the SVF approach, seed-needy households are identified through a community-based targeting approach. Communities are sensitized and organized in subvillage committees (SVCs). Each SVC develops criteria for nominating the most needy households, which differ from region to region but generally exclude households having other sources of income, an employed member, seed and food grains, more than five goats, sheep or cattle.

The targeted number of households in eastern Kenya in 2001 requiring seeds (based on the CSPD approach) and the neediest households that received seeds through SVF are presented in Table 3.

Seed procurement

In the CSPD approach, certified seed is acquired from seed companies through a standard tendering process that can be lengthy and bureaucratic. In this process the implementing agency, basically the government district agricultural staff, decides on crop types, varieties and amounts, based on the agroclimatic conditions of the district and the degree of disaster. Before a tendering process is started, each disaster-affected district presents its seed requirement to the Ministry of Agriculture headquarters. These requests are usually specific to crop and varieties. The most commonly requested crops are maize, sorghum, beans, cowpea, green gram and pigeonpea. Two to three varieties per crop are usually requested, from among the varieties that have been developed and released in the ASALs by research institutions. The requests are summarized for all districts and orders are given to the seed company that wins the bid.

In the majority of cases of seed procurement through CSPD, the Kenya Seed Company, a government parastatal, has been awarded contracts to supply emergency seed to the government. More often than not, the crops requested by the districts are not among those stocked by the seed companies.

In SVF, a two to three-day survey is conducted to find whether grains are available to be sold as seed at the divisional and locational levels. Farmers, traders, stockists and seed companies are then informed that they should take seed or grains to selected sites where seed fairs will take place on a specified day. Seed is then exchanged for vouchers or cash.

Seed distribution

In CSPD, seed from seed companies' warehouses is transported to the disaster-affected district and then to the divisions and locations where the seed is to be distributed to the beneficiaries. The distribution is done by the chief and assistant chiefs at the locational level. In most cases, there is a blanket distribution of seed. Everybody who comes is provided with seed irrespective of whether he/she needs the seed. In the distribution by SVF approach, the beneficiaries are given vouchers, which they exchange for crops and varieties of their choice and in amounts desired.

Lessons learned

Availability and access to seed/grain

Seed procurement through CSPD has always been based on the fact that, after drought, seed is not available and not accessible to farmers within their localities. The local seed assessment survey conducted before the onset of rains in 2000 and 2001 revealed that:

1. Even after four seasons of drought, such as those experienced from the short rains of 1998, long rains of 1999, short rains of 1999 and long rains of 2000 in eastern Kenya, farmers still had seed of their landraces.

2. Some local traders in market centres had between 5 and 30 bags of grains (seed) of millets, cowpea, green gram and Dolichos lablab that they were waiting to sell to farmers at planting time. (This is a most striking finding.) Most of the grains were purchased from farmers at harvesting time during the short rains of 1998 and the long rains of 1999.

Thus, grains of suitable crop species and varieties were usually available within the locations. However, farmers could not access them due to lack of purchasing power.

Efficiency and effectiveness (simplicity, timeliness and cost-effectiveness)

Through the SVF approach, CRS Kenya was able to target farm families and distribute seed to more than 8 000 farm families in the year 2000 and 30 278 in the year 2001, within a three-week period before the onset of the rains. In 2000, 275 seed vendors distributed seeds at 14 seed fairs, while in 2001, 2 169 seed vendors participated in 51 seed fairs. Farm families accessed seed of their choice prior to the onset of the rains. In the CSPD approach, seed reached farmers in late November and most of the farmers did not plant this seed (nor did farmers have a choice regarding the crops/varieties received).

The scale of operation of the SVF approach clearly indicates that seed fairs are simple to implement and can be planned and implemented within a short period. The administrative and logistical burdens, such as transport and procurement procedures, are reduced once sensitization is properly done.

In the SVF approach, farmers and traders transport seed to seed fair sites using bicycles, ox-carts and even donkeys. Elaborate transport arrangements that usually delay seed distribution are not required. Furthermore, when seed companies and research institutions are involved, they make their own transport arrangements to and from the seed fairs sites without involving second parties.

Seed vendors

A total of 275 seed vendors (sellers) participated in the 14 seed fairs conducted during the short rains of 2000 in three districts (Embu, Tharaka and Mbeere). Most of these were farmers and traders from local markets. During the short rains of 2001, 2 169 seed vendors participated, of which 55.6 percent were women, 42.3 percent men and the remaining 2.1 percent were composed of seed companies: Western Seed Company, seed stockists (Mkulima in Machakos and Makueni) and KARI (Katumani Seed Unit). In Mbeere and Tharaka, Catholic Church institutions such as Kamurugu (Embu) and Gatunga parish (Tharaka) also brought and sold seed to beneficiaries. Although most of the seed vendors were traders in local markets, who are also local grain seed suppliers, a significant number of them were farmers who brought their own farm-saved seeds.

The SVF approach provides a forum for seed companies to advertise themselves. It also has enabled research institutions to promote improved dryland crop varieties in pre-release stages.

Crops and varieties

The involvement of many seed vendors in SVF made it possible for a number of crops and varieties to be exchanged. In 2000, some 34 varieties of seven crops were brought to the seed fairs. In 2001, seeds of 43 easily recognizable varieties of 10 dryland crop species (Table 4) exchanged hands during the seed fairs in eastern Kenya. Within each of the dryland crops, there were a number of varieties/cultivars that farmers could easily differentiate by local names and preferred characteristics. Among the crops, beans had the largest number of varieties (eight), followed by cowpea with seven varieties. Farmers could also choose among six sorghum, five maize, four each of pigeonpea and Dolichos lablab, three each of mung beans and pearl millets, two chickpea and one proso millet varieties. Finger and proso millets, Dolichos lablab and chickpeas were also brought to the seed fairs but in smaller quantities.

Different locally adapted crops and crop varieties, some of which were not available in the formal seed sector but are important to food security in drought-prone areas, were brought for sale. This underscored the desired biodiversity in crops and varieties as well as farmers' preference for the various crops in each location.

The seed fairs provide an opportunity for local seed vendors and seed-needy farmers to interact. It also provides an opportunity to gather information on the kinds of crops and varieties available for sale and on the varied preferences of farmers. In this way, the SVF approach strengthens rather than undermines the operation of the local seed systems.

Through the CSPD, the main crops delivered are usually maize, bean and sorghum. In some seasons, smaller amounts of drought-tolerant crops such as cowpea, sorghum, millet and pigeonpea are delivered. The number of varieties delivered through CSPD depends on the released varieties. For maize, only two varieties (KCB and DLC) have been released for arid and semi-arid areas. However, when it comes to emergency seed delivery, hybrid varieties such as H511, H512, H612 and H 624, meant for medium and high-potential areas, are also distributed by the seed companies. Only three varieties of sorghum (Serena, Seredo and KARI Mtama 1) and three of beans have been bred and released for arid and semiarid areas. However, at no time do the seed companies have sufficient quantities for these varieties to cover emergency distribution needs.

TABLE 4
Crop and variety diversity of seed exchanged for vouchers in the seven districts in Eastern Province of Kenya during seed fairs in 2000 and 2001

Source: CRS/FAO, 2001 - 2002.

More often, the crops requested by the districts through CSPD are usually not among those stocked by these seed companies. Thus, while the districts' agricultural teams may request seed of different varieties of sorghum, millets, cowpea, pigeonpea, green gram or bean, it is hybrid and composite maize and vegetable seed that are routinely supplied.

Quantity of seed

In SVF, the amounts of seed the voucher holders purchased and took home depended on the prices of the grains (seed) in the fairs. With the voucher value of 700 Kenyan Shillings (K Sh) (about US$10) given to farmers during the seed fairs in 2001, a "typical" beneficiary used 250 to purchase maize, 250 to purchase beans, 100 to purchase sorghum and millet and 100 for other grain legumes (Table 5). At an average price per kg of maize, beans, sorghum/millets and other legumes of K Sh 16, 38, 26 and 35, respectively, the beneficiaries took home about of 30 kg of grain to be planted as seed.

TABLE 5
Amount of seed that each beneficiary took home during seed fairs in 2001

Source: CRS/FAO, 2001-2002.

In the CSPD approach during 2001, the implementing agency or the Ministry of Agriculture decided the amount of seed received by households. In most districts, each household was to receive about 10 kg of maize, 5 kg of sorghum, 5 kg of beans and 2 kg of either cowpea, green gram or pigeonpea, making a total of 22 kg. However, the amount of seed actually received by the households depended on what was supplied to the location or division and the number of households at a distribution point. During 2000 and 2001, most farmers receiving seed through the CSPD approach took home, on average, 3 to 5 kg of seed to plant (Bashir, 2001).

Formal vs informal seed systems

During the planning of seed fairs, it was thought that farmers would use the opportunity to purchase certified seed. However, farmers preferred to purchase local grains to plant as seed. Although this choice was largely attributed to the differences in prices between local grains and certified seed, interviews with farmers gave additional revealing insights:

• Farmers trusted the local grains they could see (not packaged materials). They also trusted grain traders who purchased grains from them at harvest.

• Farmers trusted grains from their neighbours' crops, i.e. for the crops they had seen growing in the field and admired.

• Farmers had had negative past experiences with seed distribution. In the previous emergency seed operations, farmers complained that seeds brought from outside (purchased from seed companies located in Rift Valley Province) and distributed to them were not suitable for their areas and never performed well.

• Most of the certified seed brought to the seed fairs was from Western Seed Company. Farmers believed that the company was located in Western Province and its seeds were therefore not suitable for Eastern Province, where the seed fairs were conducted.

With competent, experienced and proactive management, the SVF approach can provide farm families with a choice between farmer and formal seed, as well as access to small quantities of seed of new varieties.

Cost-effectiveness

In addition to accessing seed on time, through the SVF approach farmers were able to access more seed, that is between 20 and 30 kg of grains in exchange for the K Sh 600- 800 vouchers (Table 6), which would have purchased the equivalent of only 6 kg of commercial maize seed.

TABLE 6
Cost analysis of seed distributed in 2000 and 2001 using CSPD and SVF

Source: FAO, 2002.

Flexibility

The SVF methodology presents a "level playing field" upon which the commercial seed sector (seed companies and stockist) and the farmer seed system (farmers and market traders) can compete. However, the playing field can be easily tilted in favour of the commercial sector by lecturing farmers on the superiority of commercial seed of improved varieties. It can also be tilted towards the farmer seed system by encouraging voucher holders to buy locally so as to prevent the proceeds from the sale leaving the community.

Community empowerment

As opposed to blanket distribution of seed aid, in which farmers have no options, the seed fairs empowered the farmers to choose crops they wanted to plant and the amount of each crop. For example, farmers opted to purchase more grain legumes by using the seed vouchers, as the price of the grain legumes was much higher in the local markets than that of cereals.

During the seed fairs, the communities themselves identified their representatives to participate in seed inspection and setting of seed prices with the vendors. By inspecting seed themselves, the communities learned about seed quality.

Strengthening local seed supply systems

For dryland crops such as pigeonpea, cowpea, mung bean, common beans, sorghum, millets, open-pollinated maize, and tuber and root crops, farmers in dryland areas obtain their seeds from local sources, especially own-saved seeds; social networks (relatives and other neighbouring farmers); and local traders (shopkeepers and open-air market grain traders). Local traders play a key role in rural communities by purchasing grain at harvest, storing it and selling it back later to the same farmers either as food or as seed at planting time. These already existing seed institutions are critical to the livelihoods of the marginalized and poor households in the supply of food and seed grains. The desire by humanitarian agencies to assist vulnerable households in response to disaster can undermine these institutions and interfere with the coping mechanisms and livelihoods of the people in the community.

The SVF approach, through its participatory nature in implementation and other subsidiary activities, contributes to capacity building. It also contributes more to maintenance of existing biodiversity compared with the CSPD system. However, when it comes to biodiversity enhancement, the CSPD system contributes more because the nature of its operational features enables it to bring into the community more new varieties compared with the SVF system. As mentioned above, the CSPD system contributes less to local capacity building because most of its activities are restricted within the official circles, which comprise either government officials or employees of relief agencies.

References

Bashir, M. 2002. Monitoring and evaluation of seed distribution during 2000 short rains season. Presented at the workshop Strengthening Emergency Seed Support in Kenya, Machakos, Kenya, 12-14 June 2002, organized by CRS/FAO.

Braun, H.M. 1982. The 1982 agroclimatic zone classification map. Nairobi, Ministry of Agriculture, Republic of Kenya.

CRS/FAO. 2001-2002. Emergency seed distribution by voucher system for the "long"rains in eastern Kenya. Project OSRO/KEN/101/UK. Progress Reports, 2001-2002. Nairobi, Catholic Relief Services and FAO.

FAO. 2002. Comparative financial analysis of the seed vouchers and fairs scheme. Rome/Nairobi.

Government of Kenya. 1997-2001. District development plans: thirteen districts of eastern Kenya. Nairobi, Ministry of Planning and National Development. Republic of Kenya.

Otado, J.A. & Ingosi, A.K. 2002. Emergency seed aid distribution in Kenya: experiences from the Ministry of Agriculture and Rural Development. Presented at the workshop Strengthening Emergency Seed Support in Kenya, Machakos, Kenya, 12-14 June 2002, organized by CRS/FAO.

RETHINKING SEED RELIEF INTERVENTIONS: AN ETHIOPIAN CASE STUDY - George K. Mburathi, Christel Bultman, Felix Mathenge, Osman Ibrahim

George K. Mburathi, Christel Bultman, Felix Mathenge, Osman Ibrahim
FAO Ethiopia

Introduction

Ethiopia has experienced "drought emergencies" of varying intensity every year for at least the last 18 years. At the time of writing (2003), the country is dealing with the aftermath of yet another food crisis. Providing farmers with seeds, in the context of the prevailing drought conditions, has been generally considered as appropriate because (1) it directly provides assistance to targeted beneficiaries that allows for a sustainable recovery of their livelihoods; (2) the international community is open to interventions that break the cycle of massive food deliveries now underway into Ethiopia; and (3) seed interventions have been requested as a priority by the Government of Ethiopia at federal, regional and local levels.

FAO, together with many NGOs and international organizations, is involved in large-scale seed relief and other emergency interventions in order to assist rural families to recover. It is anticipated that seed assistance will be on the agenda of international humanitarian organizations throughout the coming decade. The frequency of agricultural crises, their repeated occurrence and the growing questioning of the effectiveness of seed relief interventions are sufficient reasons to start rethinking conventional approaches to seed relief.

The increasing acknowledgement that seed relief interventions by themselves cannot address the underlying causes of the emergency is one of the more pressing issues in current emergency assistance programmes. Furthermore, when seed assistance is not properly applied, there is a danger of undermining local coping mechanisms and demotivating target communities away from recovery and development. It is estimated that smallholders produce 90 percent of Ethiopia's crop output. For centuries, Ethiopian farmers (smallholders) have used local varieties/landraces from the informal seed sector for food production. Some 85 percent of Ethiopian smallholders are believed to be dependent on the informal seed sector for their seed sources. The formal seed sector, which produces certified seed of improved varieties of cereals, legumes and oilseed crops, is dominated by the Ethiopian Seed Enterprise (ESE). ESE produces an average of 12 percent of the total potential seed demand, estimated to be around 100 000 tonnes, with only a small part reaching the small-scale farmer and the rest going to large-scale farms (Tafesse, 1997).

In this paper, FAO in Ethiopia reviews three years of experience with seed assistance programmes, from 2001 to 2003, and explores the advantages and disadvantages of four different approaches of seed procurement and distribution. The analysis serves as a stimulus to rethink conventional practices of seed interventions. In the first section, the methodology of assessments that guide choice of intervention is explained. In subsequent sections, the interventions themselves are compared against the requirements of the country, the government's commitment to the various approaches and several parameters of "good practice". The final two sections comment on the evolution of practice and recommendations for the future.

Assessments

The biggest constraint to a timely and effective international community response in Ethiopia has been the lack of a systematic countrywide assessment of seed security and requirements. The FAO/World Food Programme (WFP) crop assessment provides data on crop production, but specific information about access and availability of seed is lacking. Because the Ethiopian seed system is area specific, as in other countries, and informal seed marketing systems are dominant, a seed assessment would be costly, time-consuming and require a complex set-up.

Because of the lack of such an assessment, the FAO/WFP crop assessments for 2003, several joint food security assessments (WFP, Government of Ethiopia, NGOs) and other information were taken as a starting point for selection of target areas. The various regional and zonal governments^[8] in Ethiopia provided FAO with further information on priority areas for seed interventions. Because of the time constraint and the magnitude of the intervention, it was not possible to do thorough assessments in all areas. The exceptions to this were the following cases: a participatory rural assessment (PRA) in December 2002 in Tigray Region; and in the case of East Hararge, an assessment by an ongoing FAO project on strengthening seed supply systems at the local level. Market surveys and seed requirement assessments were also carried out in the woredas selected for the seed fair approach in South Wollo Zone. In the Southern Nation, Nationalities and Peoples Region, data from a previous PRA (carried out in 2001) were used to identify preferences of farmers and traditional famine-strategic crops.

In all other cases, FAO had to rely on seed assessments and action plans carried out and developed by government experts, in target areas, often through local committees (which included representatives of the beneficiary community). Where possible, although with limited coverage, data were verified through discussions with farmers.

During field visits to all woredas, the seed requirements, beneficiary lists and action plans as indicated by the woreda governments were thoroughly discussed with the agricultural experts and fine-tuned with FAO policies and technical guidelines.

Implementation methodologies

Access versus availability

The framework developed to understand seed security better is composed of three parameters: availability (adequate quantity of seed); access (adequate income or other resources to acquire seeds); and utilization (acceptable quality of preferred crop varieties) (Remington, 1998; Remington et al., 2002).

The principle underlying the examination of different seed intervention methodologies in this paper is the acknowledgement of access as opposed to availability being the main (but definitely not the only) problem in the current seed crisis in Ethiopia. Lack of seed availability in times of food shortage is not a given fact. Even during a poor harvest, farmers will try to retain some seed needed for the following season. It has to be said, however, that seed retained after a drought year is often of poor quality. Also, if farmers anticipate that they can rely on extension packages or seed relief from the government or NGOs, they will not hold on to their seeds. For certain crops such as sorghum, quantities of seed required are relatively small, so that it is easier for farmers to retain their adapted varieties. Nevertheless, this is not true for all crop varieties and in some areas of Ethiopia with generally poor harvests, this seed is of poor quality and often not kept for planting.

Rural seed systems are remarkably resilient. Seed can usually be acquired through friends, relatives, as a gift or by barter. Seed can also be purchased from local markets and petty traders, as a cash purchase or a loan to be repaid after the harvest. Usually, relatively better-off farmers do not suffer from drought as much as vulnerable farmers. The problem during and after (repeated) drought is that the seed system comes under stress and the problem for farmers becomes access to seeds rather than a total absence of seeds in the woreda.

Seed categories

The ways in which both accessibility and availability should be addressed depend on the category of seed being considered. In the projects implemented, and those in the process of being implemented by FAO in Ethiopia, the following seed categories should be kept in mind.

• Certified seeds of improved varieties. This category of seed is produced from varieties registered in the Crop Variety Register with quality control provided by National Agricultural Input Authority (NAIA). Certified seeds are expensive and in most cases have not penetrated to small rural farmers. Procurement and distribution of this category of seed, in an emergency assistance environment, should be considered only in cases of total crop failure where availability is a problem or in cases where they are accepted by farmers as preferred varieties.

• Local seeds/landraces of locally adapted varieties. These seed varieties are not certified by NAIA and are produced and maintained at local levels by farmers and local institutions. They constitute the bulk of seeds used by small farmers. They are popular with farmers for their adaptability to local conditions of low input agricultural systems as well as suitability for local food habits. This category of seed has both farmers' and FAO's preference for seed relief in most areas in Ethiopia.

Other categories of seed are vegetatively-propagated planting materials and exotic vegetable seeds. Since these seed types will not be discussed further in this paper, they are not included here.

FAO seed procurement and distribution approaches in Ethiopia

FAO has had experience with four methodologies of seed relief interventions in the country during the period 2001-2003: a) local, committee-based procurement and distribution; b) seed fairs and vouchers; c) national tender through FAO in Ethiopia; d) tender through FAO headquarters. The methodologies of the four approaches are outlined below.

Local, committee-based procurement and distribution

A local committee-based procurement and distribution programme is a general methodology for seed intervention at the peasant association level. It is responsible for the local purchase of seed for redistribution in times of seed crises. In the case of emergency seed assistance, such distribution contributes to the recovery of farmers' livelihoods after drought or other types of shocks, but the methodology can also be applied to other types of interventions. The problems identified in the community are taken as a starting point for decision-making on types of inputs to be distributed and other types of assistance required. The community itself, through its representatives and with the assistance of the peasant association, development agents^[9] and woreda agricultural experts, is responsible for the problem analysis, decision-making, implementation and follow-up of the project. FAO guides and monitors the process closely and assists the committees and government authorities that are responsible for intervention for an optimal project impact.

The main element of the process is the formation and operation of several committees at the community level: the seed survey committees, beneficiary selection committees, and procurement and distribution committees. The committees are responsible for the planning and actual implementation of the project.

The committees often consist of the same members because their activities are interlinked. The output of the seed survey committee is an analysis of seed requirements and availability and an action plan to assist the community. The analysis is used as a basis for the activities of the procurement and distribution committee that is the actual implementer of the formulated action plan. The beneficiary selection committee selects the beneficiaries for assistance according to the same action plan and makes sure that the procurement and distribution committee distributes the seeds to the right beneficiaries. For a more detailed overview of responsibilities of the various committees, see the Appendix.

This approach has been applied by FAO in Ethiopia during the implementation of several projects in the period 2001-2003.

• Under a project for assistance to internally displaced persons (IDPs) in 2001, 17 936 beneficiaries were provided with 382.9 tonnes of locally purchased seeds of finger millet, sorghum, wheat and barley (all local varieties).

• The approach was further applied by FAO during 2002 in a project for provision of seeds to drought-affected farmers and support to irrigation. A total of 19 797 beneficiaries were provided with 766 tonnes of seeds of local varieties of wheat, barley, teff, maize, sorghum and chickpea.

• In the 2003 crisis, the approach will be used again for two FAO Technical Cooperation Programme projects for farmers in need of seed. Assuming that future distributions can be carried out through the woredas, 282 000 beneficiary households will be assisted with approximately 7 300 tonnes of seeds (mainly wheat, teff, pulses, sorghum, maize, barley, potato and finger millet).

For all targeted zones, the improved seeds were bought from recognized seed growers^[10] such as the Swedish International Development Agency seed multiplication schemes and cooperatives. Local seeds were acquired from individual seed traders and farmers. In some cases, improved seeds were purchased from the Ethiopian Seed Enterprise. However, the bulk of the seeds were bought through local purchases.

Seed vouchers and seed fairs approach

Seed fairs are markets where beneficiary households purchase seeds through a voucher system. Catholic Relief Services (CRS) introduced the seed fair approach in Ethiopia and FAO is collaborating with this NGO in a joint seed fair programme in South Wollo Zone in Amhara Region. Seed fairs are organized on a specific day and location, announced in advance and attended by registered farmers and seed vendors/suppliers, who are also preregistered. Seed fairs aim to address the access problem by providing farmers with a means of purchasing seed/crop varieties in quantities of their choice and also to strengthen the local seed supply system and promote linkages and information sharing among farmers (CRS, 2002). FAO conducted one seed fair in South Wollo Zone in the highlands of Ethiopia as a pilot. After selection of the woredas, an awareness-raising/training session was organized at the zonal level with involvement of all stakeholders. Subsequently, seed fair committees were established at the woreda/PA level; these consisted of local experts, government authorities, and representatives of beneficiaries so as to involve all stakeholders. Requirements, seed availability and local marketing channels were assessed on the basis of questionnaires that had been agreed upon during an initial workshop by these committees in the selected woredas. As in the local committee-based procurement and distribution approach, selection of beneficiaries and suppliers was carried out by the committees under the supervision of the woreda authorities. A quality control was carried out under the supervision of FAO. The seed fair itself involved the following activities:

• registration of beneficiaries and suppliers (committee/FAO/CRS/local NGO);

• final quality control of seeds (FAO);

• coupon distribution (FAO/CRS/local NGO);

• coupon/seed trade - monitoring of rules (FAO/CRS/local NGO);

• evaluation questionnaires with beneficiaries and suppliers (CRS);

• redeeming of coupons by suppliers (FAO/woreda authorities).

At the seed fair, 2 000 beneficiaries were assisted with a total cost of US$49 601. The amount of seeds provided depends on the preferences of the beneficiaries. In the seed fair that already took place, five different seed varieties of sorghum, various varieties of maize and sesame were sold, with a total of approximately 3 500 kg.

National tender by FAO in Ethiopia with selected registered seed suppliers

The procedure of conducting a national tender in Ethiopia will be explained on the basis of a project that was carried out by FAO in East Hararge, in the far east of the country. FAO staff were and continue to be present at the project site. The aim of the project was to support resource-poor farmers in order to restore their seed security for the forthcoming cropping season so that crops can be produced, leading to improved household nutrition status and reduced dependency on external food aid. The project was fully embedded in a coordination platform of implementing organizations, chaired by the Disaster Preparedness and Prevention Commission, which is the coordinating body of the Government of Ethiopia in emergencies. The total seed needs were determined through woreda authorities and local experts. Because of the lack of an institutionalized mechanism to do this, seed needs are generally calculated based on crop production assessments. A seed procurement committee was established at the zonal level, chaired by FAO. Beneficiaries were selected by the woreda/PA authorities. The procedures for procurement were as follows:

• preparation, announcement and collection of public national bids based on specific requirements;

• bids screening;

• quality control test at recognized national seed laboratory according to standards set by the Ministry of Agriculture;

• procurement of seeds through formal seed supply system with reasonable prices and assured seed quality standard.

Seed was distributed under an established agreement between FAO and a qualified NGO or woreda agricultural office in the target area. The agreement clearly specified the type of crops and varieties, qualities and quantities to be distributed; the procurement, handling and distribution costs; and the obligations of the implementing partner with regards to reporting. The distribution of seeds took place through the committee at the woreda level.

In the project, 11 500 beneficiary households were provided with 441.5 tonnes of wheat, pulses, sorghum, maize (five varieties), barley and potato seeds. The seeds were purchased from the Ethiopian Seed Enterprise and one other private seed enterprise.

National tender by FAO headquarters

Under existing FAO rules and regulations, procurement of seeds through issuing of a local purchase order by any FAO Country Representation with a monetary value above US$25 000 is not allowed. In such cases, FAO headquarters undertakes the procurement of the seeds through a national tender with registered seed suppliers, recognized by the National Registering Authority. Under normal circumstances, after assessment of requirements and selection of target woredas, the FAO Country Representation initiates the procurement process by submitting to FAO headquarters detailed technical specifications of what needs to be purchased. FAO headquarters handles everything from that point until the seeds are delivered to the stated destinations in the country. The selection of beneficiaries and distribution of procured seeds rests with the project implementation authorities - in this case zonal and woreda authorities.

During 2001, FAO headquarters procured seeds for wheat (two varieties of improved seeds) and maize (two varieties of improved seeds) with national bidding, totalling 315 tonnes for a project to assist drought-affected farmers. The seeds were distributed to 3 133 beneficiaries. The purchase order was issued to a private seed enterprise. The company delivered the first 150 tonnes of wheat seeds at the beginning of June 2001. The remaining quantity was transported to the region during the subsequent weeks. The purchase through the national tender was accomplished in line with the initial plan. However, problems arose when the beneficiaries and local authorities were not satisfied with the crop varieties and the seed quality.

Coverage of needs

In December 2002 a joint Government-UN appeal for emergency assistance was issued in which the Ministry of Agriculture requested US$15 213 687 for the provision of local and improved seeds to be used in the 2003 planting season. Of this total request, FAO in Ethiopia covers US$3 112 145 in its current projects (Table 1).

In general, one can state that FAO is covering 20 percent of the total of the other agencies' seed requests in the country, with 78 percent of the total request actually achieved (although these percentages are highly simplified compared with the real situation).

Government commitment to the various approaches

The Government of Ethiopia has established a general policy that has agriculture as the engine for development. Attaining food security is the nucleus of the agricultural programme. In order to achieve this, "there would have to be major improvements in traditional agricultural practices; here, the use of improved seeds would be crucial" (EARO, 1999, p. 19).

Nevertheless, the Ministry of Agriculture acknowledges the fact that distributing improved seeds as an emergency intervention may not be appropriate, especially in marginal environments and when not combined with inputs such as fertilizers. In 2003, the Government of Ethiopia is providing funds to assist farmers to obtain seeds. For this purpose, the Ministry of Agriculture has delegated responsibility to the zone/woreda level to purchase local seeds. For improved seeds, the policy is to obtain seeds mainly from the Ethiopian Seed Enterprise. Depending on the level of authority, the Ministry of Agriculture in general favours the local purchase of local varieties and national tendering for purchasing of improved varieties for seed relief.

Lessons learned

In this section, the different methodologies/projects will be discussed on the basis of the following parameters: cost-effectiveness; utilization (quality and preference); strengthening or weakening of local seed systems; gender sensitivity; capacity building (of woreda government and farmers); timeliness; and logistical capacity of the implementing organization. The cost-effectiveness of the four methodologies will be compared in the next section. In the following section the remaining parameters will be discussed.

TABLE 1
Overview of requirements and contributions for seed emergency interventions in May 2003 (in US$)

Cost-effectiveness

Developing a method for measuring the cost-effectiveness of seed relief is difficult because of the differences in farm size and crops grown and also raises the question of whether a capacity-building parameter should be included. Nevertheless, possible measures of cost-effectiveness used in this paper are total costs per beneficiary and total costs per hectare. Since Ethiopia is in the middle of the cropping season, it is not possible at the time of writing to measure the actual crop production of the ongoing projects. This is unfortunate since it would provide another measure of cost-effectiveness.

The quantity of seed each beneficiary should receive in an emergency is a question that is under discussion in Ethiopia. The answer depends largely on the area a family is cultivating and the crop variety, which can vary in each project area. For example, the beneficiaries of the seed fair were provided with seeds for 1 ha, while the beneficiaries of the local committee-based procurement and distribution programme only received seeds for 0.25 ha. These decisions were made because the average size of land that can be cultivated by a single household is significantly different in the two areas. For interventions other than the seed fair, it was difficult to gather data distinguished by crop type since each farmer made their own choices. Therefore, the comparison here is made using costs per hectare.

The data below, for Tables 2 and 3, are derived from a project in six woredas in Oromiya Region (community-based seed supply); the seed fairs in two woredas of South Wollo Zone; and the project in East Hararge in two woredas in Oromiya (national tender), respectively. The data for a tender through FAO headquarters are not available in Ethiopia and are therefore not reflected in the table. Presumably, the numbers will more or less correspond with those of the tendering through FAO in Ethiopia.

Table 2 presents information on the cost per beneficiary, as well as overall costs of the projects, total numbers of beneficiaries and numbers of households involved.

Table 3 draws from the same data, but from the slightly different perspective of costs per hectare.

Seed cost refers to the value of the seeds provided to the farmer. While in the local committee-based approach and the national tender many crops were distributed, the predominant crop in the seed fair was sorghum. This results in a relatively low price of the seeds in the seed fair, because sorghum is relatively inexpensive.

TABLE 2
Comparison of three approaches: total costs, scale of service and costs per beneficiary (in US$)

TABLE 3
Costs of different approaches per hectare (in US$)

*Awareness raising and instructions for the committee-based programme and national tender are included in the transport and staff costs as, besides travel and staff, no other costs are incurred.

Note: LCPD = Local committee-based procurement and distribution.

In interpreting the tables, we need to recognize that the scale of the project is very different in all three approaches. In the case of the seed fair, some of the costs per hectare would be reduced if greater quantities of seeds were distributed. In addition, the costs of printing the coupons and awareness raising means this approach will remain significantly more expensive than the other approaches (at least in the start-up stages) but offers the advantage that it supports the local seed system. The staff costs of seed fairs and other approaches are not directly comparable, because the biggest seed fair is still a relatively small-scale operation with relatively large staff requirements compared with those of the other approaches. However, these costs could be reduced in future seed fairs, with savings in awareness-raising strategies and printing of materials and a reduction in numbers of staff involved (i.e. less monitoring will be required as familiarity with the approach increases).

Overall, Tables 2 and 3 show that the local committee-based procurement and distribution approach is the most efficient, with the other two incurring significantly higher costs. Despite the fact that the seed fair was dominated by relatively less expensive seed, this approach fares the worst in terms of cost-efficiency. The labour and awareness-raising requirements of seed fairs suggest it will remain relatively expensive.

Comparison of approaches for the remaining criteria

After the comparison of costs for all approaches in the previous section, Table 4 gives an overview of the remaining parameters for comparison for each of the approaches.

Farmers' opinions were directly sought in comparing the quality and kinds of seed distributed under the different approaches, through a series of surveys among the beneficiaries. In the case of the local committee-based approach, 20 beneficiaries were randomly selected from a woreda covered by the ongoing project in Amhara Region. The survey participants were, in general, satisfied with the crops distributed (19 out of 20 were satisfied with both the quality and variety of seeds). Nevertheless, a long list of seeds were mentioned that they would have preferred for additional distribution, but were not made available to them during the distribution (red sorghum, faba bean, teff, field pea, bean, lentils, chickpea, legumes and maize). Eleven beneficiaries purchased additional seeds in the market and only one beneficiary purchased the seed distributed to him and sold another type of seed.

In the case of the seed vouchers and fairs, 20 beneficiaries were randomly selected from Kalu woreda. The overwhelming majority of interviewed farmers (19 out of 20) were satisfied with the quality of the seeds in the seed fair. The fact that they were able to select their own seed was clearly mentioned by all farmers as an advantage of the seed fair. Farmers only mentioned two types of seed that were missing at the seed fair (teff and chickpea). Four of the interviewed farmers bought additional seeds and no one sold seed to purchase something else. Many farmers mentioned the social control mechanism in the woreda as a guarantee for good quality seed.

TABLE 4
An overview of various parameters for each approach

In the case of the national tender through FAO in Ethiopia, it became clear from the survey that, although farmers appreciated the seeds they received, the farmers receiving the local sorghum variety had reservations about the quality (some impurity and less germination than the improved seed). All interviewed beneficiaries (13) said they were satisfied with the crop varieties they received. The large majority of the farmers in the sample purchased additional seeds from the market. Interviewed beneficiaries also reported being satisfied with the procedures because the project was participatory from the start.

In the 2001 assistance to IDPs, used as an example here because this approach was not applied in 2003, the regional authorities gave low quality and late delivery of the supplied seeds as reasons for refusing complete acceptance of the improved wheat seeds. Although the region eventually received the remaining 220 tonnes of the improved wheat seeds from the supplier, the optimal planting time for wheat had almost elapsed before the issue of delivery was finally resolved. In the end, an agreement was reached between the regional authorities and the project office to keep the remaining 220 tonnes of improved wheat seeds for distribution during the next cropping season (Meher, 2002). Since the project was completed some time ago, no survey data is available.

Evolution of approaches

As mentioned above, the vast majority of the seeds obtained through the traditional seed system in Ethiopia come from local marketing or through informal trade and exchange. Although providing farmers with improved seeds can be useful, this approach must be handled with care for several technical and non-technical reasons (among others, the use of second and third-generation seeds, the difficulties farmers experience in obtaining fertilizers and the disruption of local seed marketing). Therefore, FAO in Ethiopia focuses on local or improved but locally adapted seeds in its current emergency interventions. An exception is made for crops that are not present in the local market, such as vegetables.

In general, experience has shown that the easiest method to obtain locally adapted and preferred seeds is to purchase from local traders, farmers and small-scale local seed enterprises or cooperatives. In the two approaches focusing on national tender processes, the local seed system is not included. As explained in the introduction, neglecting local seed systems can be harmful not only in terms of effectiveness of the intervention, but also in relation to long-term consequences. In addition to this, the tender system at the national level is not as appropriate as the other approaches in terms of gender sensitivity and capacity building. It is also more difficult to deliver to farmers the seed of their preference, although the example from East Hararge showed that beneficiaries will appreciate quality seeds from the formal system, as long it has their preference. In addition, it needs to be stated that many potential problems can be avoided through beneficiary participation.

A national tender through FAO in Ethiopia is a process that is reasonably manageable, although the factor of time is a debatable issue. Another advantage of national bidding is that additional quality tests can easily be carried out under the national authority that has registered the bidding suppliers.

Both the local committee-based and seed fair approaches are preferable because they facilitate better options in easily acquiring crops and varieties that are adapted to each locality and preferred by each community; they also address the access issue, build capacity at least to a certain degree and strengthen the local seed system. Even (locally adapted) improved varieties can be acquired from local seed production schemes within the project area. The committees involved in the local woreda distributions ensure to a certain extent that local preferences are reflected in the types of seed distributed, while in the seed fair approach, this issue is addressed directly, since farmers choose the seeds they purchase with vouchers.

The seed fair approach, though comparable in costs to the conventional methodology based on national tender, appears to be very expensive and time-consuming compared with local committee-based procurement and distribution, but "wins" in terms of utilization, gender sensitivity, and capacity building. Therefore, a way should be found to combine advantages of the seed fair with the cost- and time-effectiveness of local purchase and distribution.

General reflections

• In view of the lack of a systematic methodology for assessing seed needs in the country as a sound basis for mounting emergency interventions, deliberate efforts should be made to investigate how to move beyond emergency seed supply to provide alternative forms of assistance that more accurately reflect the needs of farmers affected by disaster. The development and use of suitable tools for seed needs assessment could be one priority area for FAO and other humanitarian agencies involved in seed aid. The development of suitable seed system profiles could be a good starting point.

• Once the need for seed intervention has been established and targeting prioritized, the most efficient and effective procurement and delivery mechanism should be applied to enhance the impact on target beneficiaries. The existing rules and regulations that may not be conducive to emergency assistance response requirements in terms of relevant delivery parameters such as timeliness, cost-effectiveness, quality of inputs, gender sensitivity and others need to be revisited in order to tailor them to the requirements of an emergency environment.

• As much as possible, efforts should be made to link and support, from the outset, emergency seed assistance with activities that promote self-reliance and capacity building in seed self-sufficiency on the part of target beneficiaries to reduce the need for recurring seed assistance demands in the future. Pilot farmer-based seed production activities and associated farmer training in such activities are viable options.

• In addition, mechanisms should be put in place in a seed relief intervention that trigger more sustainable outcomes in addition to the intervention itself. These might include revolving funds, in kind, linked to the development of seed banks. Another possibility is the multiplication of improved seeds in the woredas by selected farmers with a guaranteed premium price when selling seed to suppliers.

• Due to the recurring nature of droughts in Ethiopia, water harvesting, storage and utilization in drought-prone areas where water for crop and livestock production as well as for domestic use is in short supply should be promoted and supported in order to cushion beneficiaries and enhance the impact of seed assistance since such water resources can be used in many ways to promote food security.

References

CRS. 2002. Seed vouchers and fairs: a manual for seed-based agricultural recovery in Africa. Catholic Relief Services, developed in collaboration with International Crops Research Institute for the Semi-Arid Tropics and Overseas Development Institute. Nairobi.

EARO (Ethiopian Agricultural Research Organization). 1999. An economic development strategy for Ethiopia. Addis Ababa.

Tafesse, K. 1997. Towards seed industry development in Ethiopia. In Proceedings: International Workshop on Seed Security for Food Security, pp. 201-206. FAO, Rome and Accademia dei Georgofili, Florence, Italy.

Remington, T. 1998. Increasing the effectiveness of emergency seed aid programs in enhancing seed security in the Greater Horn of Africa: a project proposal. First submitted to USAID/OFDA, 9 September 1998.

Remington, T., Maroko, J., Walsh, S., Omanga, P. and Charles, E. 2002. Getting off the seeds and tools treadmill with CRS seed vouchers and fairs. Disasters, 26(4): 302-315.

APPENDIX

Responsibilities of committees under local committee-based procurement and distribution

1. Seed survey committee (community level-PA)

Tasks of the committee:

• To adjust the template seed survey to their specific situation.

• To agree on the sample size and sampling method of the survey.

• To carry out the survey in their respective PA.

• To analyse the survey results and make the Community Seed Supply Action Plan.

• To report the outcomes of the survey and the Action Plan through the woreda agricultural office to FAO at the agreed time.

• To adjust the Action Plan according to comments and advice from FAO.

Reporting to FAO:

• The agricultural office clears the outcomes of the survey and the requirements and sends them for clearance and advice to FAO Representation.

2. Beneficiary selection committee (community level-PA)

Tasks of the committee:

• To select the beneficiaries in their respective PA according to the criteria set in the briefing session and according to the Action Plan.

• To organize a plenary meeting with the community, to review the beneficiaries and decide on the final beneficiary list.

• At the time of inputs distribution, the committee will ensure that only selected beneficiaries are given the seed inputs.

• To ensure accountability of distributed seed inputs through certification of collection (signatures/fingerprints) by beneficiaries on prescribed inputs collection sheets.

Reporting to FAO:

• Reporting to FAO takes place when the FAO regional coordinators^[11] collect the lists of beneficiaries with their signatures for receipt of the seeds, bearing the stamp of the woreda council or agricultural office.

3. Seed procurement and distribution committee (woreda and community level)

Tasks of the committee:

• To review availability of required inputs in the local markets as reported by the seed survey committee and decide on preferable sources for purchase of inputs (seeds and/or others). The inputs can be obtained through the following categories:

  a) local markets (farmers), in the respective PA, in other PAs or in other woredas;

  b) local traders or cooperatives;

  c) local seed enterprises, in the respective woreda, zone or region;

  d) federal seed enterprises.

• To analyse availability of inputs through the identified sources.

• To identify sources for purchase of inputs and analyse the market situation, to be reported to FAO.

• To ascertain the quality of seeds to be purchased.

• To follow the preparations for procurement.

• To purchase the seeds.

• To arrange transport from the seed supplier to the PA or woreda and assure adequate storage.

• To set an appropriate date for distribution and announce the date to the beneficiaries.

• To distribute the seeds and receive signatures of all beneficiaries.

• To account for all finances and reports to FAO.

Reporting to FAO:

• The committee, through the woreda project implementation authority, reports to FAO potential sources for seed purchase, including a market analysis or tendering process.

• Receipts will be sent to FAO to account for all expenditures.