An FAO/WFP Crop and Food Supply Assessment Mission visited Ethiopia from 4 November to 1 December 1995 to estimate the production of the 1995 Meher (main) season cereal and pulse crops, forecast 1995/96 Belg (secondary) season production, estimate national food requirements in 1996 and assess the food aid needs for that year. The Mission members visited all the zones in the four main crop producing regions of the country, but were unable to visit regions in the far west or the far east, whic h are largely pastoralist or hunter/gatherer domains, contributing little to the national harvest.

This year the Mission was joined by observers from the major donors, including representatives from the European Union (EU), United States Agency for International Development (USAID), Christian Relief Development Association (CRDA), Save the Child ren Fund (SCF, UK), the International Federation of Red Cross and Red Crescent Societies (IFRC) and CARE International, who travelled with the teams and monitored activities during the extensive field visits.

Data obtained from the zonal offices of the Ministry of Agriculture (MoA) and the Commission for Disaster Preparedness and Prevention (CDPP) were cross checked with information received from national and regional officers, woreda (district) level o fficials, NGOs, farmers, traders and donor representations. The teams undertook crop inspections and field, household and market surveys to confirm the information received. In addition, the Mission was provided with official population statistics, a comprehensive bibliography of reports produced during the year by governmental and agency groups, including rainfall data prepared by the EU's Food Security Unit, the returns of national market surveys and mid-term reviews on crop condition. The amalgamation of inter alia, the Ministry of Coffee and Tea Development and the Ministry of Agriculture (MoA) improved access to data in several of the zones in Oromia and SEPAR regions. In addition to this year's data, the Mission also received up dated data on the 1994 harvest based on MoA conducted surveys. As a result, comparisons between 1994/95 and 1995/96 have been based on the most recent returns, not last year's Mission's figures, which have since been superseded.

The Mission forecasts a 1995/96 production of cereals and pulses of 9.4 million tons, comprising 9.1 million tons of Meher and 335 000 tons of Belg crops which is some 13 percent above the recomputed data for 1994/95. Timely and well-distributed ra ins throughout the previous Belg and Meher rainy season not only encouraged planting and supported the satisfactory development of the crop in most zones, but also provided the draught animal force with the optimal conditions for cultivation.

Further, the effects of the liberalization of the grain trade encouraged farmers to make best use of the opportunity resulting in a national area increase of 5 percent. An improved distribution of fertilizers, which increased by 16 percent and virt ually no migratory pest problems completed a combination of conditions which have not been as good for many years.

On the negative side, non-migratory pests have still taken their regular toll of produce, poor access to draught animals reduced both the timeliness and quality of cultivations and planting, input supply remained well below demand and weed competit ion was also favoured by well distributed rains. In several areas, premature rain-stop denied the potential promised earlier but was not enough to cause concern except in chronic drought-prone lowland areas. With the exception of woredas in Centra l and Eastern Tigray, North and South Wollo, North Omo and West Hararghe, conditions may be regarded as comparatively comfortable. Storage losses are expected to be higher in the vulnerable crops due to increases in temporary home-made silos, longe r period of storage envisaged and the presence of already noted pests and disease.

Other crops have also benefited from the well-distributed rains. The coffee crop is reported to be good and root crops are yielding well. However, the main alternative staple, enset, though comparatively immune to seasonal fluctuations is a cause f or concern due to the continuing spread of bacterial wilt, which is now said to be reducing noticeably the yield in some areas.

Cereal prices, which had been comparatively stable for all crops throughout the year until August, were noted by the Mission to be falling abruptly in most areas where the harvest was underway due to the combined effects of the arrival of new grain s and the off-loading of old stocks by traders and farmers in anticipation of the new supply.

Given the Government's estimate of population for the coming year of 56.67 million persons, the level of carryover stocks and the forecast production for the coming year, the Mission considers that no cereal imports will be necessary in 1996, other than limited amounts of food aid already pledged. However, despite a comfortable supply situation overall, a significant proportion of the population will be in food deficit due to displacement, underproduction or lack of access/entitlement to foo d supplies. The Mission estimates that about 3 million persons in the country will require emergency food assistance of 291 000 tons during 1996. In addition, some 100 000 tons of cereals for the build up of the Emergency Food Security Reserve and 30 000 tons for food aid stocks will be required in 1996. These requirements are expected to be covered by already committed pledges, a sharp drawdown of in-country food aid stocks and donor-supported local purchases estimated at 235 000 tons. Shou ld these local purchases not materialize, additional food aid imports will be required to meet the emergency needs.


Rainfall during the 1995 cropping season

The onset of the Belg rains was by and large timely, with delays reported only in scattered northern zones. As a result, cultivation for the Meher season proceeded on time prompting farmers to sow more long cycle crops, a practice which increased p roportionally from the north to the south of the country reflecting the rainfall pattern. In most of the zones in the four main crop growing regions (Oromia, Amhara, SEPAR and Tigray), germination rains were also timely and generally well-distribut ed geographically, resulting in a much reduced need to replant compared to 1994. Indeed, the necessity to replant due to drought was largely restricted to lowland zones in West Hararghe (Oromia). Similarly, interference from excess rain at planting time was restricted to localized areas at higher altitudes. Waterlogging delayed planting and combined with hailstorms ultimately stunted growth and reduced production in some high altitude areas of Amhara and in parts of central and eastern Tigra y.

In general, the Meher season rains followed the long term average distribution pattern for most of the growing season in most areas, with the exception of the West Hararghe lowlands where crucial deficits were noted in May and September. Unfortunat ely, premature rain-stop in some of the highlands of Amhara and in central and eastern Tigray reduced grainfill in the areas with soils of low water holding capacity. In southern regions, the autumn rains have begun well, encouraging the establishm ent of late planted root crops as well as cereals. Notwithstanding the existence of chronic drought-prone lowland areas, the rainfall this season was probably nearer to optimum than has been experienced for many years in both quantity and distribut ion.

Area Planted

The area planted to cereals and pulses increased in all regions, except Tigray, compared to the latest MoA figures for the actual 1994/95 season collected by the Mission. All cereal crops show increases in area (in Tigray at the expense of pulses) with greater increases noted in the southern half of the country. As the Belg rains were timely, draught oxen entered the main rainy season in good condition and the liberalized grain trade plus universal access to local seeds encouraged farmers to take advantage of the opportunity the rains afforded to expand activities wheresoever possible, resulting in an estimated increase in area planted to grains of 6 percent.


The yields observed by the Mission were greater than 1994/95 yields due to the advantageous distribution of rains in time and space and the comparative absence of migratory pests. However, due to the wide range of agroecological and edaphic conditi ons which can change dramatically from place to place within districts, yield figures collected by the Mission on a zonal basis in several instances masked significant variations between highland and lowland areas which, in turn, created pockets of concern in areas of comparative comfort.

As indicated above, migratory pest attacks were minimal. Only one woreda in eastern Tigray was affected by locusts and a quelea quelea infestation of sorghum in East Hararghe was speedily controlled by aerial spraying organized by the MoA.

Inadequate access to animal/tractor power was also considered to be responsible for delayed onset of ploughing, reduced number of cultivations for farmers with only one or no oxen, delayed sowing, reduced germination potential and increased need fo r weeding later in the year.

Further, although the national distribution of fertilizer went up by 16 percent from 206 000 tons to 240 000 tons, thereby, at an annual rate of 1 quintal (100 kg) per hectare, theoretically increasing the production potential of another 340 000 he ctares, problems regarding timely delivery to districts away from the main roads and timely access to credit to farmers with outstanding loans, reduced its availability for use on the long cycle crops. The introduction this year of a second distrib ution network organized by a second commercial source of supply may help to improve distribution in coming years.

Apart from signalling the use of fertilizer, the comparatively high yields of barley and wheat in Arsi and Shoa also reflect the relatively easy access to improved seeds in such areas. Elsewhere, the availability of improved seeds was restricted to pilot extension packages disseminated via the Global 2000 or the National Extension Programmes. Such pilot packages, though impressively numerous and well distributed across the country, are limited to 0.5 hectare plots so do not yet make a signif icant contribution to the national harvest. They have, however, generated a great deal of interest through the yields obtained on farmers fields, which ranged from 2.5 tons to 8 tons per hectare according to location and seed type used.

Non-migratory pests such as stalk borer (maize), shootfly (teff), leaf-eating ladybird (teff), termites (all cereals), wild animals in the forest and near-forest areas, and the bush cricket in Wollo, were noted as problems by the Mission. Stalk bor ers on maize and sorghum were the most ubiquitous, being reported in every region. Unfortunately, access of farmers to pesticides was limited by both availability and price except in areas where specific programmes were underway to provide plant pr otection facilities for cash crops.

Despite the effects of the factors identified above, all of which are normally present but usually effectively masked by the stronger adverse effects of poorly distributed rainfall or migratory pests, the Mission forecasts that yields this year wil l be higher for all crops in all regions.

Cereal and Pulse Production Forecast

With a few exceptions, harvesting was well advanced in the regions during the visits, enabling the Mission to observe yields and to hold confirmatory discussions with farmers. The Mission forecasts the 1995 cereal and pulse production of the Meher season at 9.1 million tons, of which 38 percent will be maize and sorghum. Eighty-one percent of the national production of cereals and pulses is expected to be produced in the Oromia and Amhara regions, which is a similar proportion to last year, reflecting both their potential and their access to inputs.

The forecast production is 27 percent higher than last year's Mission forecasts but is only 15 percent higher than the revised MoA figure for 1994/95 determined by the zonal agricultural offices after the harvest was completed. These revised data, which were collected by the Mission, indicate a 1994 Meher grain output of 7.9 million tons against 7.3 million tons forecast by the 1994 Mission. The revised estimate reflects in part better reporting conditions brought about by the amalgamation o f Ministries and the unification of agricultural services in coffee growing areas particularly in the SEPAR region. The new data also helps to explain why, despite a shortfall of imports, no critical situations developed over the past year.

Table 1 provides estimates of areas and forecast production for all cereal crops and pulses by region. A comparison with last year's post-harvest review figures is given in Table 2. In all cases, areas used are based on MoA zonal office figures col lected during the Mission. Yields for the current harvest have been adjusted on the basis of field inspections at farm and woreda level.

Table 1 - Area ('000 ha ) and Production ('000 tons) of cereals and pulses - 1995/96 Meher Season



AMHARAArea1 015.7552430.7220.522.30.1








ADDIS ABABAArea5.20.33.5




ETHIOPIAArea2 361.21 402.21 402.4404.555.60.1
Prod.1 533.41 345.81 567.7278.252.70.1

Table 1 (cont.)

AMHARAArea279.35663 086.6681.93 768.5
Prod.363.6514.62 509.2396.92 906.1
OROMIAArea853.36062.83 736.74314 167.7
Prod. 1 141650.11.44 130.5265.64 396.1
SEPARArea342.1104.41 035.5182.81 218.3
Prod.455.598.21 056.1117.51 173.6
DIRE DAWAArea0.23.844
ADDIS ABABAArea92.811.8
ETHIOPIAArea 1 547.21 530.231.18 734.51 335.410 070.7
Prod.217.31 497.612.88 305.6794.49 100.0
Note: Totals derived from unrounded data.

Table 2 - Cereals and Pulses Production : Comparison of 1994/95 and 1995/96 Meher Season






AMHARA1994/952 910.22 215.6690.93823 601.12 597.6

1995/963 086.62 509.2681.9396.93 768.52 906.1
OROMIA1994/953 428.53 461.6427.2242.73 855.73 704.3

1995/963 736.74 130.5431265.64 167.74 396.1

SEPAR1994/951 010.7928.3162.494.11 173.11 022.4

1995/961 035.51 056.1182.8117.51 218.31 173.6

ETHIOPIA1994/958 212.87 160.81 352.4744.49 565.27 905.2
1/1995/968 734.58 305.61 335.4794.410 070.09 100.0

1/ 1994/95 - MOA updated post-harvest figures corrected by Mission from zonal agricultural officers.

Note: Totals derived from unrounded data.


Given the well-distributed Belg rains, livestock of all classes entered the main rainy season in good condition. This state was sustained throughout the main season, particularly in the less densely populated areas of the south and west, where stan ding hay, browse and plenty of cereal/pulse by-products and a timely beginning to the late autumn rain suggest that no shortages of fodder are likely in the next few months.

In the pastoralist areas, production from the livestock industry is not expected to have altered significantly this year and so is not expected to change the demand for cereals and pulses in such areas.

In the densely populated areas with associated restrictions on areas of grazing and browse, livestock condition was noticeably poorer. In many areas the grazing was finished and as arable by-products had yet to become available, there was a hungry- gap in which stock were mobilizing their body reserves. The only obvious exception was noticed in the highlands of Hararghe where farm systems have evolved including the fattening of draught oxen on hand-fed thinnings from the late sown sorghum cro p. Such an integrated system, which may have application elsewhere, was producing animals in extremely good condition and providing the farmers or the area with a substantial financial return when the animals were sold just before harvest.

The only serious animal diseases reported to the Mission were an outbreak of rinderpest in Eastern Tigray, which instigated a vaccination campaign in the zone, and endemic presence of trypanosomiasis in the lowland river valley areas. Information g athered during the Mission with regard to trypanosomiasis suggests that as more riverine lowland comes under the plough, the tse-tse fly is moving into the higher forests. Farmers faced with the dilemma of using draught animals under such condition s are finding their own solutions, including short term (seasonal) purchase and resale of oxen in neighbouring tse-tse free areas, but the long term solution is felt by the Mission to be appropriately scaled mechanization on lease/lend systems back ed by adequate servicing.

Livestock prices were reported to be either stable or rising in all regions, with seven year old draught oxen in good condition fetching 800-1 000 Birr according to locations and number of market presentations.


As noted last year, crops other than cereals and pulses contributing to household nutrition economy vary from north to south and from east to west. The oilseed crops, most important in the north and east, such as neug, flax, linseed, sesame and gro undnut, showed no noticeable increase in area, but are likely to have better yields than last year. Similarly, the coffee crop which is predominantly grown in the south and west, is expected to be good this year, reflecting the biennial fluctuation s of coffee production as well as the effect of well-distributed rainfall in coffee growing areas. Another major seasonal cash crop, chatt was said to be less productive than last year but as the prices of the leaf are inversely proportional to pro duction and the market is guaranteed, chatt producers were expecting good returns.


Studies during the year have confirmed the importance of enset as the staple food of around 13 percent of the population. However, as the processed commodity, kucho, is presently not easily transferable from its site of production and processing, i t is not viewed by the Mission as an exchangeable food item. As such, although the consumption of it and other root crops grown in the south, have the effect of reducing the national per caput cereal consumption, actual increases in production of e nset are not capable of reducing national cereal consumption below a certain threshold. Conversely, factors negatively affecting enset production or processing will immediately increase cereal demand in the south and concomitantly increase the nati onal per caput consumption figure. Of immediate concern in this regard is the spread of bacterial wilt in enset. From observations made during the Mission it is apparent that the disease is spreading, and is now estimated to be reducing production by 10 percent in some areas. There appears to have been no progress made in either identifying the pathogen or in finding any form of plant protection other than sanitization, and farmers are becoming increasingly worried. If the infection continue s unabated at the same rate, cereal demand could increase by 50 000 tons per year in the near future.

The root crops of importance, Irish potatoes, sweet potatoes and taro benefited from well distributed rains and good harvests are expected in both Oromia and SEPAR, where their production is significant for both cash generation and household nutrit ion.



Tigray is the northernmost region of Ethiopia bordering Sudan and Eritrea. With a cultivated area of some 0.84 million hectares it contributes around 5 percent of the national grain harvest in a good year, but because of its high population density it is traditionally a food deficit area. This year, the Belg rains were slightly late, but once started were reliable in both intensity and distribution. A similar pattern was observed with the onset of the main season rains, however in both the e astern and central zones short term droughts in May and June affected seed set and premature rain-stop reduced grain filling below the potential of the crop in areas with soils with low waterholding capacity. Although the total area planted was sim ilar to last year, an increase in cereal areas was noted due to a switch from pulses because of the promising beginning to the season and the combined effects of the increased availability of fertilizer and encouraging market conditions for cereals .

Fortunately, the threat of locusts from Sudan and Eritrea identified in September failed to materialize except in one woreda in the eastern zone, due to prompt action along the borders and advantageous wind changes. As no other migratory pests chal lenged production, the only noted yield reductions due to insects came from regular non-migratory pests such as stalk borer (maize and sorghum), shoot fly (teff) and grasshoppers(teff).

The combined effect of the above factors has resulted in a forecast harvest of cereals and pulses which is 12 percent greater than last year's harvest (Mission adjusted Zonal Agricultural Office figures), of which 83 percent is expected to come fro m the western and southern zones.


The Afar region, located in the northeast of Ethiopia is essentially a low-lying plateau accommodating camel and small ruminant pastoralists. Given rainfall usually below the requirements for crop development, production is obtained only on the bas is of irrigation. Presently, few foodcrop producing schemes exist to the extent that the MoA (Addis Ababa) had no available figures for foodcrop production in the region for this year, but as last year's figures only amounted to a total of 106 tons , this is not considered by the Mission to be a significant omission. Reports from other observers suggest that the livestock in the region are in good condition and rainfall has supported the development of adequate browse and grazing.


The region, sub-divided into ten administrative zones, accounts for 37 percent of Ethiopia's cropped land and around 32 percent of its grain production. Of the region's approximately 4.6 million hectares of arable land, about 3.7 million hectares i s currently under cultivation. Altitudes in the region range from about 700 metres to well over 4 000 metres above sea level, with intensive areas of crop land at altitudes between 2 000 and 3 000 metres. Traditionally, the region, by increasing al titude, is divided into five agro-ecological zones: Bereha, Kola, Woina Dega, Dega and Wurch, each with its own rainfall pattern, agricultural production system and exposure to crop failures. Dominant soil types are black vertisols (43 percent) and black non-vertisols (27 percent).

On the whole, the Meher rains started at the expected time and were thereafter well distributed, except in some low areas where the rainfall was frequently erratic and inadequate for achieving good yields. Reports of excessive rains were received f rom parts of the higher altitude zones, which resulted in waterlogging, delays in land cultivation and planting, stunted growth of the crops, and serious infestations of weeds. Unfortunately, in most of the region the relatively favourable rainfall conditions came to an untimely end in late August or early September, causing water stress in growing and maturing crops, especially on the lighter soils and on degraded terrain where soil is shallow and water-holding capacity is consequently grea tly reduced. The crops most affected were the long cycle crops (maize and sorghum) and late planted pulses.

As regards incidence of pests and diseases, conditions during the current season have been much better than last year, as no migratory pests were noted. Serious localized damage was caused by stalk borers, Wollo bush crickets, aphids, African bollw orm, grasshoppers and barley flies. Plant diseases such as smut on sorghum and rusts on wheat were also observed to have reduced yields.

Notwithstanding a rather early end of the rains, East Gojam, West Gojam and Agawawie zones enjoyed a good agricultural year. The rainfall pattern encouraged good land preparation which was followed by good vegetative growth and fruit setting. Combi ned with a somewhat improved fertilizer and seed supply, and a low incidence of crop pests and diseases, above-average crops are expected. As a result, the current crop is estimated to be the best of the last three years.

In North and South Gondar, the crop initially performed better than last year, with very encouraging prospects until the end of October. From then, however, due to the early termination of the rains, yield estimates (especially of late sown crops l ike pulses) required a drastic downward revision. This year's production in the zone ranges from a good average in the highlands to extensive failures in some lowlands.

According to the Mission estimate the forecast harvest for Amhara region's 1995/96 Meher season is 2.91 million tons of cereals and pulses which is 12 percent greater than last year's Mission revised figures.

While this year's relatively good growing conditions have thus led to certain production increases, the constraints the agricultural sector faces in drawing the full benefits from such improved conditions should be stressed. Particularly serious in this respect is the lack of agricultural traction (oxen) which prevents many farmers from working their land and planting their crops during the relatively short period when conditions are optimal. The generally low level of agricultural technolog y and low soil fertility, as a result of rampant land degradation, are two other important considerations.


Oromia, the largest region, consists of a "T" shaped land mass in central and southern Ethiopia, extending from the western to eastern to the southern borders of the country, encompassing highland plateaux, lowland river valleys and a wide variety of agro-ecological zones in between. The region is divided into twelve zones of varying sizes and production potential.

Throughout the region, administrative changes have improved local methods of data compilation through the amalgamation of the previously separate offices of the Ministries of Natural Resources, Coffee and Tea, and Agriculture. In addition, improved security in the western zones (Wellega) and no population movements improved both data collection as well as production possibilities.

Throughout the region, the Belg rains began on time and were well distributed resulting in reasonable Belg harvests in the limited number (four) of Belg producing zones. In non-Belg producing areas, more than adequate browse and grazing for oxen an d good cultivating conditions resulted in a significant increase in area planted to maize and sorghum. Fortunately, in the main production areas of West Shoa and Arsi, West and East Wellega and Jimma, the promise of the Belg season was sustained by well distributed rainfall throughout the Meher season, which, though less than the long term average in most cases, was sufficient to guarantee seed set and reasonable grainfill for all cereal crops.

Lowland areas in the region were less fortunate, particularly in West Hararghe and North Shoa, where all lowland woredas have reported crop failures to varying degrees compared to adjacent middleland and highland woredas where conditions have been most favourable and good crops were observed.

Zonal agricultural office reports, confirmed by field inspections and farmer interviews throughout the region, indicate a year virtually free from migratory pests. The only exception noted was an attack by quelea quelea on sorghum of the southern w oredas of East Hararghe, which was contained by aerial spraying organized by the zonal agricultural office. As elsewhere, non-migratory pests were considered to be a major problem throughout the region. Important localized pests were termites in We st Wellega and wild animals in all forested areas whose presence is a continuous threat to crop production from seed emergence to harvesting.

An estimated 131 000 tons of fertilizer was distributed, which was some 9 percent greater than last year. Distribution difficulties resulting in late delivery in remoter areas have still to be solved as do problems regarding access to credit for fa rmers who previously belonged to bankrupt peasant associations.

Given the favourable conditions early in the season, replanting was unnecessary except in lowland woredas in West Hararghe, and as local seed supply was good the planted area was extended. However, improved seed was available only for limited crops and in limited zones. If the interest created by the success of the SG2000 and National Extension packages is to be sustained and the formula adopted on a satisfactory scale in the region, the distribution of the essential ingredients, seeds, herb icides and pesticides, will have to be improved.

As a result of the favourable conditions in most of the region, a harvest of 4.39 million tons of cereals and pulses is forecast from the 4.16 million hectares cultivated. This is an improvement of 18 percent over last year's MoA revised figures fr om an 8 percent increase in area.


Located in the south and south western area of Ethiopia, the Southern Ethiopia People's Administrative Region (SEPAR) encompasses 11 zones and five special woredas. All the zones and woredas experience Belg and Meher rains. This year the Belg rains conformed to the expected pattern, beginning in February and continuing through to May. Thereafter in most zones the Meher season began in June without any significant break. The rainfall of the Meher season then continued to fulfil expectations t hroughout the season except for localized reductions in June and July, which were generally compensation for later in the season but may have had local effects on seed set.

Due to the good initial conditions, no replanting was necessary. However, the early nature of the season meant that many peasants planted the long cycle crops without fertilizer as either it was not available or they had not yet paid off loans from the previous season and were unable to obtain new loans in time for a timely delivery.

Generally speaking, the region is not a cereal consuming area. Enset is the preferred staple. The continuing spread of bacterial wilt of enset, noted by the Mission, is therefore a serious cause for concern and detracts from the overall feeling of well-being engendered by the good coffee crop in what is one of the main coffee growing regions. In addition, cereal districts, such as Konso, were seen to be less fortunate due to the patchy nature of the rainfall received and their greater depend ence on cereal crops.

With regard to other factors affecting yield of cereals and pulses, pest damage reports conform to the situation in other regions. There were no migratory pest attacks during the year but non-migratory pests such as stalk borer and leaf-eating lady birds were commonplace. As reported last year, the farmers' greatest cause for concern this year with regard to pests was crop loss to wild animals and the debilitating effect on the farming community of the need to guard cereal crops from seeding to harvest against such pests.

Given the almost contiguous nature of the Belg and Meher rains, the Mission observed greater problems from weed competition and anticipated higher pre/post-harvest losses as crops were left standing for longer in the field before harvest, to the ex tent that weevils were seen to be attacking the standing crop. Storage losses will also be greater as crops were being brought into the local silos already infected with smuts, cob rot and weevils.

The condition of livestock throughout the region was observed by the Mission to be good, except in densely populated districts such as Wolaita where shortage of forage is a perennial problem. No major disease outbreaks were reported and livestock p rices were said to be stable and equivalent to those reported elsewhere in the country at 800 to 1 000 Birr for a seven year old draught oxen in good condition.

According to Mission estimates, the forecast harvest of cereals and pulses for the SEPAR 1995/965 main season is expected to be 1.17 million tons from 1.2 million hectares which is 32 percent more than last year due to improved yields and the inclu sion of areas hitherto unreported.

Benshangul, Gambella, Addis Ababa

Production figures from the remaining regions (except Harer, which is included in the Oromia data) were provided by the MoA's head office in Addis Ababa. The three regions are not thought to contribute more than 75 000 tons of cereals and pulses. G ood opportunities for increased production exist, especially in Gambella, but infrastructure development and low population density is presently limiting the full exploitation of the region.


The Belg harvest traditionally encompasses those crops which are grown and harvested during the Belg rains which usually extend from the first dekad of February until the end of May. Other crops planted during the Belg, but harvested during the lat ter part of the Meher season, such as maize and sorghum have been included with the meher crops as the produce will contribute to 1995/96 consumption availabilities.

The 1995 Belg rains were timely, resulting in a greater planted area. Production reported from the ten main Belg producing woredas throughout the country was estimated by MoA at 395 400 tons, which was substantially higher than the harvests of the past two years. Following recent estimates of the Belg crop for the past three year produced by MoA, and the lack of recent estimates from the Central Statistics Authority, the 1996 cereal and pulse harvest during the Belg season has been forecast by the Mission at an average level of 335 000 tons of which 25 000 tons are expected to be pulses. This figure, based on MoA's production estimates is lower than the average Belg crop derived from CSA series.


Prices of the five main cereal crops remained comparatively stable during the main growing season across the country until August/September, when they began to fall, due to traders selling stocks in anticipation of a good harvest. Since October, th e decrease has become more pronounced particularly for cereals such as maize which had by then reached a more advanced stage of harvest. In the Addis Ababa market, the maize price in October fell by 16 percent from the previous month. During the Mi ssion, prices as low as 50 Birr per quintal (100 kg) were recorded for maize compared with the lowest price of 91 Birr per quintal at the same time last year in the same districts. There are, however, wide zonal price differences which reflect prod uction patterns. Poor communications, particularly in areas away from the main roads, exacerbate the differences, reducing the prices of surplus commodities and increasing the prices of cereals not produced in such areas.


The 1996 projected balance between grain availability and utilization is shown in Table 3.

Cereal and pulse production in marketing year 1996 (January-December) is calculated at 9.4 million tons, with the main Meher crop estimated at 9.1 million tons and the secondary Belg crop, to be harvested before next May, provisionally forecast at 335 000 tons. This production forecast should be revised after the Belg harvest and the current supply/demand forecast adjusted accordingly.

Foodgrain stocks in the country by 31 December 1995 are projected at about 465 000 tons. Relief stocks, amounting to 141 000 tons, are somewhat below the previous year's high level, while those of the Ethiopian Grain Trading Enterprise (EGTE), esti mated at 224 000 tons, are several times higher than in 1994 following the good harvest of 1995 and late arrivals of programmed food aid. Stocks of the old crop in the hands of farmers and private traders are assumed to be about 100 000 tons. These carryover stocks do not include the country's Emergency Food Security Reserve (EFSR), estimated at 208 000 tons of grains.

The Mission forecast that closing stocks at the end of 1996 will remain at the relatively comfortable level of 1995. Despite an above-average production this year, no overall build-up of stocks is anticipated as the country is expected to cover a l arge part of its consumption requirements from the domestic availabilities. However, it is forecast that cereal stocks will change hands among the economic agents. The Emergency Food Security Reserve is expected to increase to 307 000 tons as a res ult of recommended 100 000 tons of local purchases to be made with donors support. This volume of security reserve will cover about two weeks' food requirements of the population in case of emergencies. Stocks held by EGTE are forecast to increase slightly from the current levels, reflecting heavy intervention purchases, as well as late arrivals of 1995 programmed food aid pledges in 1996. Stocks held at household level and by private traders are also foreseen to raise following this year's bumper crop and the increasing effects of the market liberalization. By contrast, the Mission anticipated a substantial drawdown of relief stocks during the 1996 marketing year (January-December) resulting from the 1996 emergency food aid distribut ion programme, coupled with the adoption of local purchasing as the main means of supply by donors next year and the policy of maintaining minimum relief stocks other than those of the Emergency Food Security Reserve (EFSR). Stocks of relief food a id by the end of 1996 are thus anticipated to be the minimum necessary to cover contingencies derived from the need of timely mobilization of the food assistance at regional level.

Food consumption of grains has again been calculated using an average 123 kg of cereals and 12 kg of pulses per caput per annum. It should be underlined that these figures refer to the long term consumption of grains and not to any recommended nutr itional levels. It is assumed that the consumption of other foods, namely enset, root crops, milk and meat provide the difference needed to maintain a calorie intake necessary to meet at least the minimum nutritional requirement. However, this is n ot to say that all sections of the population receive the amount of grain needed to meet the standard. Food deficit areas exist due to localized production failures, concentrations of displaced persons and the capability of areas to achieve househo ld self-sufficiency, even in good years, due to structural and climatic limitations and a lack of alternative employment.

The revised population figure for use in the supply/demand analysis was supplied to the Mission by the Central Statistics Agency. At 56.67 million, it is 270 000 greater than the figure used last year and conforms to official predictions for the Et hiopian calendar year "1987/88".

Therefore, the 1996 food consumption requirements amount to 7.65 million tons of cereals and pulses which is only 0.05 million tons more than the estimated consumption last year. Given the variability of the population statistics quoted by differen t Ministries, this figure should be viewed with some caution until verified by the release of the returns from the recent national population census expected in January 1996.

It is important to note that if the UN population growth rate of 2.98 percent per annum were applied to the figure used by last year's Mission, the estimated food requirement would be 190 000 tons more.

With regard to feed consumption, large livestock is mainly fed by grazing, browse and crop residues in the traditional sector and by feedstuffs constituted from mill wastes in the modern sector. The country's 3.5 million pack animals receive irregu lar grain supplements which could amount to 50 000 tons a year when grains are available.

However, discussions held by the Mission with livestock specialists, feedstuff manufacturers and poultry producers indicate that it is the poultry industry (traditional and modern) that is by far the greatest grain based feedstuff user. Recent loca l studies funded by FAO suggest a modern poultry population of 2 million birds consuming at least 73 000 tons of grain per annum. Therefore, if the traditional poultry population, which is estimated at a minimum of 20 million birds, each consuming a variable and unknown quantity of grains, is taken into consideration, it is most unlikely that combined animal feed use is less than 150 000 tons per annum. It should be noted that this could very well be reduced in a poor year when supplies are expensive and grain feeding in the traditional sector is cut back to near zero.

Other uses including seeds, brewing and post-harvest losses have previously been estimated at 15 percent. The Mission feels that this is an underestimate. Firstly, virtually all finger millet is likely to be used for traditional brewing and a subst antial proportion of barley in Arsi goes for malting, suggesting a total brewing demand of some 500 000 tons. Secondly, seeding rates of maize (30 kg/ha), sorghum (20 kg/ha), wheat and barley (100 kg/ha), teff and millet (20 kg/ha) would suggest a seed demand of 407 000 tons. Thirdly, although teff and to a certain extent sorghum, local wheat and local barley are comparatively immune to post-harvest losses, losses of maize are very high. This year's good harvest means that farmers will be st oring their grains longer, both for home consumption and to take advantage of the trading possibilities afforded by liberalization. If storage losses are calculated according to the formula teff x 1 percent; sorghum/wheat/barley x 5 percent; maize x 20 percent, the overall loss estimate is likely to be 7.5 percent. As a result of the foregoing, the revised percentage for other uses/losses for the 1995/96 Meher season is 7.5 percent.

Cross-border exports in 1996 are expected to increase as a result of the good harvest and the decline in cereal prices, coupled with a reduced cereal crop in neighbouring Eritrea. Based on current indications of grain movement, the Mission estimate s that the pull of better prices will affect a much greater movement to Eritrea for all cereals than in previous years. This will be increased if the Sudan/Eritrea border remains closed and official cross-border transfers of cereals is limited. In addition, exports are also likely from the southern surplus regions to south Sudan, where production continues to be affected by the civil war. The anticipated exports from Ethiopia, therefore, are assumed to increase to some 80 000 tons.

Cereal import requirements in 1996
Apart from the expected arrivals of 80 000 tons of food aid already pledged (45 000 tons of relief food aid and 34 000 tons of programmed food aid) no additional cereals imports are forecast in marketing year 1996 (January-December). This situation at the overall level masks the fact that 291 000 tons of emergency food aid will still be required to meet the food requirements of the most vulnerable population. In addition, some 100 000 tons of cereals for the build-up of the Emergency Food Se curity Reserve and 30 000 tons for food aid stocks, will be required in 1996. As indicated in table 4 below, these requirements are expected to be covered this year with expected arrivals of emergency food aid already pledged, a sharp draw-down of carryover stocks of food aid and donor-supported local purchases of about 235 000 tons. It should be underlined that the Mission's forecast is based on the assumption that these local purchases will materialize. Should this not be the case, additio nal food aid imports will be necessary. The forecast also assumes a 1996 average Belg grain production. Given the extreme vulnerability of the Belg outturn to weather vagaries, the cereal import requirements will need to be reviewed towards the mid dle of next year when the Belg harvest is completed.

Table 3 - Cereal and Pulse Balance Sheet, 1996 (January/December) ('000 tons)




1. Domestic Availability

8 616


9 435

2. Production

8 616


9 435

- Meher 1995/96

8 306


9 100

- Belg 1996




3. Total Utilization

8 696


9 515

4. Food use

6 970


7 650

5. Feed use




6. Exports




7. Other uses

1 496


1 635

8. Import requirement




9. Food aid pledges




10 Shortfall




Table 4: Food aid balance in 1996 ('000 tons)

Food aid requirements


- Emergency food aid


- Emergency Food Security Reserve


- Emergency food aid stocks


- Programme food aid


Food aid availability


- Emergency food aid carryover stocks


- Emergency food aid pledges carryover from 1995


- Local purchases


- Programme food aid pledges carryover for 1995



Food Access Situation in the Target Areas

Although in 1996 Ethiopia will be able to cover most of its food needs on a national level, access to food continues to be a problem; an estimated 2 to 3 million people in areas of high population density, and with meagre land holdings/assets and p oor access to markets, remain vulnerable to food insecurity due to limited ability to either produce or purchase enough food. Also, despite the overall favourable production, certain areas, particularly the highlands, continued to experience crop l oss and failure.

Of particular concern are the usual deficit areas in Tigray and Wollo. Household food economy surveys indicate that, even in a normal year, relief food aid contributes up to 30 percent toward food security of the poorest households in these areas. The situation is aggravated by: 1) a second consecutive bad harvest in 1995 in some areas; and 2) localized inadequate deliveries of food aid, especially in North Wollo, which have increased the vulnerability of a large sector of the population.

Other chronically food deficit areas requiring assistance during 1996 are Wolaita and parts of West Hararge; localized food shortages also exist in other parts of Amhara, Oromia and the Southern region (SEPAR).

Food Aid Requirements

Relief/regular food aid will be required in 1996 for the following categories of beneficiaries:

- rural households experiencing crop loss and asset depletion;

- internally displaced persons and returnees from resettlement schemes;

- urban destitutes; and

- people at acute risk of food insecurity and asset depletion.

The total emergency/regular food aid required for 1996 (or cash equivalent where food aid is substituted by cash-for-work) is 290 658 tons, or a little less than three-quarters of the 1995 requirement of 420,000 tons. Table 5, 1996 Food Aid Needs, provides detail on numbers of households and amounts of food assistance by zone on which this total food aid requirement estimate is based. This total is comprised of two categories of requirements and a contingency as follows:

First Category Requirements: 181 436 tons for the most affected, food deficit zones and the 375 000 internally displaced and destitutes, including those in urban areas. Destitutes will receive free food assistance up to 135 kg per year, covering a bout 75 percent of their needs.

Second Category Requirements: 95 381 tons for beneficiaries living in less affected zones or in areas with favourable market access.

A 5 percent Contingency of 13 841 tons for unexpected deficits that may arise in pastoral and other areas that depend on the coming Belg (winter) rains.

Table 5: 1996 Food aid needs



Yearly allocation by hh (kgs)

Number 1/ of households

Total (tons)



98 300

11 796



100 000

20 000



10 000

2 000



118 000

17 700


326 300

51 496



45 500

4 095



117 700

35 310



76 800

19 200



51 700

9 306



24 000

3 600



61 700

7 404


377 400

78 915



4 444




(375 000 persons)


50 625


181 436



Yearly allocation by hh (kgs)

Number* of households

Total (tons)



100 000

4 500



148 300

8 898



19 300

2 895



64 200

7 704



246 700

29 604



13 300




18 300

1 098



54 400

4 896



20 000

1 200



8 300




51 100

4 599



128 300

7 698



144 400

12 996



66 700

3 002



18 300

2 196



31 100

2 799


1 132 700

95 381



130 811


95 381


50 625

CONTINGENCY (5 per cent)

13 841


290 658

1/ This represents the total number of households in the affected areas used as a basis for calculating the average household deficit. Food assistance only will be given to the most vulnerable of these households and the individual ration will be a ccording to need.

In light of significant surpluses existing this year in the major producing areas, donors are requested to locally purchase as much of the balance of required relief/regular food aid as appropriate in order to encourage agricultural production and marketing, taking into account cost-effectiveness concerns and local commodity availability and acceptability to beneficiaries.

In addition to the relief/regular food aid needs covered in this report, WFP requires about 70 000 tons of cereal food needs for its refugee programme in Ethiopia, which is financed through separate pledges. These needs could also be covered mostly through local purchases.

In order to avoid destabilizing the still nascent local cereal market, and considering this year's good harvest, no additional programme food aid is needed for 1996. An estimated 34 000 tons of 1995 pledges for programme food aid will arrive during 1996.

Methodology for Needs Assessment

The 1996 food aid requirements were calculated by aggregating the likely household deficits in the most critically affected zones. Household deficits were determined by considering the average agricultural production by zone, combined with the bala nce of other income sources from the household food economy (i.e., livestock, cash crops or employment), against the minimum income needed for a household to be food secure. This was assumed to be 1 100 kg in cereal equivalents for an average house hold of 5 persons, out of which 100 kg will be used for seeds and a further 200 kg to cover other essential household expenses.

The Mission estimates, that a population in excess of 3 million people will need food assistance during 1996. The total amount of food aid for 1996 ( 290 658 tons) will be regionally allocated according to the aggregate household deficits and distr ibuted to the most needy beneficiaries, selected by the local Disaster Preparedness and Prevention Committees. Since individual food availability and income is not equal, not all the households will need assistance and amongst the beneficiary famil ies, most will not need assistance over the full year, i.e. will receive monthly rations of 15 kg/head or Food For Work Rations only during the most critical period of food shortages. Therefore, between 4 and 9 monthly distributions, depending on t he food deficit of the particular areas, will have to be carried out during 1996.

Carryover Stocks

Carryover stocks for relief/regular food aid at the end of 1995, including 1995 pledges expected to arrive after the end of the year, are estimated to be 186 426 tons. First category food aid needs can be covered from these stocks, depending on sto ck location. At the end of 1996, it is estimated that an operational stock of 30 000 tons will be carried forward to the next year.

Use of Vegetable Oil

The Mission recommends a thorough review of the role of vegetable oil in the food ration. During 1995, all aid agencies faced difficulties in importing vegetable oil, resulting from official concern that distributing large quantities of relief oil would discourage local oilseed production and depress prices for butter, on which many poor rural households depend for market sales.

Supplementary Feeding

Although the need in terms of nutritionally vulnerable people continued, insufficient capacity of the Commission for Disaster Prevention and Preparedness (CDPP, formerly the Relief and Rehabilitation Commission, or RRC), and inadequate targeting co nstrained the implementation of supplementary feeding programmes in 1995; large stocks of supplementary food could not be utilized. The option to implement supplementary feeding programmes using the remaining stocks will be maintained for 1996 wher ever NGOs have the local presence and capacity, and where such programmes fulfil a clearly defined and controlled supplementary feeding function.


The overall relief food assistance programme will be coordinated by CDPP and implemented through government line ministries (e.g., MoA) and NGOs according to a well-defined division of labour.

The total amount of 1996 food aid is intended to cover the needs of the food deficit population and will also support ongoing, labour-intensive public works programmes carried out by WFP and NGOs in food deficit areas outside the target areas cover ed by the regular relief programme of the CDPP. In keeping with the policy of the Federal Democratic Republic of Ethiopia (FDRE) to use relief food assistance to support productive activities, food aid will be programmed through Employment Generati on Schemes (EGS) wherever adequate capacity exists, with free food distribution limited to vulnerable groups unable to work.

The coordination of EGS activities is the responsibility of the Disaster Prevention and Preparedness committees established up to the woreda level. The CDPP is responsible for the organization and functioning of these committees, the early warning systems and allocation and distribution of food aid and other resources. EGS are to be based on plans prepared jointly by technical staff and beneficiaries through a participatory process based on the community's priorities and assessed food insecu rity conditions; these plans should be integrated with the overall development and disaster prevention and preparedness plans of the regions, and implemented with technical support from appropriate government or NGO implementing agencies.

Duration of the food assistance and the work-output to be achieved through EGS will be determined by the overall food deficit of the particular area to be covered by food aid. Participants in EGS schemes will selected by the local Disaster Preventi on and Preparedness committees from among the most needy in the community.

Review of 1995 Food Aid Programme

Food Aid Availability in 1995

The estimated emergency/regular food aid needs for 1995 were 427 000 tons. Pledges in 1995 totalled 435 275 tons. 487 859 tons were available for distribution during 1995, including 330 640 tons of total arrivals through December 31, 1995 and 157 2 19 tons of 1994 carryover stocks.

The Emergency Food Security Reserve (EFSR) played a crucial role in enabling timely relief interventions, especially EGS, in 1995, and so contributed to disaster mitigation. On the basis of confirmed donor pledges, implementing relief agencies were able to draw stocks from the Reserve. Repayment to the EFSR was then affected upon the arrival of the expected shipments. All borrowings are expected to be repaid by the end of 1995, bringing the level of the Reserve up to approximately 208 000 to ns. Apart from allowing the provision of timely relief, a system of borrowing and rotating stocks through the EFSR should improve flexibility in food allocations and forward planning and reduce carryover stocks, as late arrivals serve to replenish the Reserve. Contributions to the EFSR in 1995 were 37 275 tons.

During 1995, donors contributed only 168 616 tons as programme food aid, against estimated needs of 375 000 tons. This shortfall in programme food aid in 1995 contributed to relatively high wheat prices during the year, since the government owned m ills had to purchase wheat from the market. This benefited local producers and markets, but reduced the access of poor urban households to an important staple food.

In general, the quantities of food available during 1995 (local production and food aid) were adequate for covering the national needs without depressing local producer incentives, as evidenced by relatively moderate price fluctuations during the y ear.


The Mission estimates that, by the end of the year, about 347 379 tons of food aid will have been distributed in 1995. This represents 81 percent of the relief/regular food aid needs estimated by the 1994 FAO/WFP Crop and Food Supply Assessment Mis sion.

The Mission encountered difficulties in determining the number of people that received food aid. Both the CDPP and NGOs collect data on quantities distributed only. Numbers of beneficiaries are calculated by dividing the total food quantity distrib uted by the planned per capita food allocation (usually 15 kg cereals per month); however, as most people reported receiving less than this amount, the actual number of beneficiaries is presumably higher than officially reported. Determining the nu mber of beneficiaries was also complicated by variations in targeting criteria and intervals between distributions.

A uniform system to record and report the actual numbers of beneficiaries receiving food aid at the time of distribution is needed. Training for more systematic monitoring and reporting by relief agencies (CDPP and NGOs), especially on amount of fo od aid distributed through EGS versus free food relief, number of beneficiaries reached and EGS work activities/outputs achieved, is also needed.

In most cases, food was distributed as a household ration. Most EGS projects paid a standard daily food ration of 3 kg for 22 workdays per month, up to 5 months.

Large regional differences in food allocations were evident again in 1995, depending on the assessed need of the region, the implementing agency, the selection criteria applied and - most importantly - the available food quantities. Total amounts d istributed per household for the year, as reported to the Mission, ranged from as low as 8 kg (2 kg per household per month for a total of 4 months) up to 540 kg (60 kg per household per month for 9 months). Where the number of family members parti cipating in EGS was not controlled, a household in some cases received food aid in excess of its food deficit. Also, more accessible regions with less severe food shortages got a disproportionately higher share of food assistance. As in the previou s year, the Mission found that a dilution of the rations had occurred in order to cover a larger number of beneficiaries.

Some parts of North Wollo, in particular, were severely undersupplied when pledges to the NGO implementing agency were delayed, causing an erosion of the coping capabilities of the needy population; urgent action is needed early in 1996 to avoid a serious nutritional problem in this area.

Beneficiaries often had to travel long distances in order to collect food. Location of distribution sites should be reviewed and food assistance should be considered to cover the cost of transport to remote regions with no road access.

E xperience with EGS

Since 1993, the FDRE has had a policy of using relief food assistance to create productive assets by providing aid to able-bodied persons only as remuneration for their participation in EGS, or labour-intensive public works such as soil and water c onservation structures, roads, ponds, clinics and schools; free food aid is limited to vulnerable groups unable to work. FDRE aims to achieve a distribution of 80 percent of relief/regular food aid through EGS and 20 percent through free relief by the end of 1998.

Actual implementation of this policy has differed from region to region. EGS activities have so far been implemented extensively in Region 1 and some parts of Regions 3, 4 and SEPAR. In general, NGOs channelled much of their food through EGS, and t he Ministry of Agriculture remained the largest implementing agency of labour-intensive conservation-related public works through the WFP-sponsored Project 2488. Planning for projects implemented under 2488 was integrated with ongoing regular devel opment programmes and considered farmers' problems, technical criteria and appropriate work standards. However, the scale of the need and the lack of adequate capacity locally has meant that a significant amount of relief continues to be distribute d the traditional way, i.e., as free handouts.

Constraints on planning and implementing EGS in 1995 were: 1) insufficient capacity; 2) timeliness of resource availability; 3) insufficient community participation in planning and selecting activities; 4) mass mobilization approach; 5) limited re sources, including food, funds for training and per diem for extension staff, monitoring and evaluation stationary supplies, and tools; 6) commodity transportation and storage difficulties; and 7) mobility of government technical support staff.

Capacity: Although the focus of the CDPP emergency appeal in 1995 was "capacity building for linking relief to development", the major constraint to EGS programmes continues to be the insufficient capacity of government and NGO implementing agenci es in design, management and quality control, including: 1) low awareness of EGS policy and responsibilities for implementing EGS; and 2) insufficient administrative and technical structures, i.e., shortages of ready EGS/relief plans with a "shelf of projects" and of qualified technical staff, and insufficient integration of EGS activities into regular line department programmes.

Strengthening and expanding existing training in participatory identification and planning methods, elaboration of work plans and beneficiary selection methods/criteria for government, UN and NGO field staff involved in EGS design and implementatio n is therefore needed. The WFP-supported Local Level Participatory Planning (LLPP) methodology for involving the community in planning, prioritizing and implementing labour-intensive public works according to set work norms and quality standards is a useful point of departure. Donors are encouraged to continue to support the Government's EGS capacity building efforts. Linkages with the UNDP training programme in support of EGS capacity building should also be explored.

Part of capacity building should include improved linkages between MoA, CDPP and NGO implementing agencies.

Timeliness: Commitments for food aid are usually available from late March onward, whereas the peak period for EGS activities occurs between January and May, during the break in agricultural activities. Untimely delivery of food aid and other nece ssary inputs (cash, equipment, materials) complicates the organization of relief interventions through EGS and delays completion of activities and payment to workers. This results in EGS activities competing with agricultural work, relief food not being available during the time of food deficit, causing asset depletion among some poor households, and participation and quality of the work being negatively affected.

Community participation in designing and managing EGS activities is key for achieving quality, maintenance and sustainability of the assets created, and for ensuring that the assets meet the communities needs. However, a recent survey of EGS implem enting capacity found that participation of communities was generally limited to prioritization of already chosen activities and location, with no direct involvement in the identification process. To achieve the necessary level of community involve ment will require investing a lot of time and human resources on the part of implementing agencies.

In Regions 1 and 3, (Tigray and parts of Wollo) labour-intensive public works are realized through mass mobilization of available manpower, whereby all able-bodied persons are required to contribute 20 days to community development activities, on a non-remunerated basis. Food aid is regarded as a support to the community, not as payment for labour per se; as such, a food for work ration is provided to the neediest of the able-bodied participants in the community work activities, and free foo d only to the most disadvantaged and vulnerable in the community that cannot work, as identified by the community itself. The intent of this policy is to maintain the community's incentive to work due to a desire for development rather than seeking a reward. This practice is only of concern when: 1) the quality of the work output suffers from work not being linked to achievement of targets and technical standards; and 2) accountability and transparency to donors for proper food use is affec ted.

Also, the obligation to do free community work, along with the other extreme of setting payment at too attractive a rate, may prevent participants from pursuing other income-generating activities and traditional coping strategies, which may create future dependency on food aid.


Nearly all farmers interviewed by the Mission confirmed that a fair targeting of food aid to the poorest of the community had taken place, and that the community as a whole works to receive food. This is a large improvement over how food had tradit ionally been distributed in the past. An evaluation of the Merti-Jeju EGS programme found that the project successfully targeted 85 percent of the most vulnerable groups with able labour for participation; participants in this category where found to work longer and more frequently than less vulnerable households. Nevertheless, about 15 percent of all households lack able labour and could never benefit from EGS activities; free relief safety net mechanisms must be maintained for this group.< /P>

Typically, EGS activities are selected by the local Disaster Preparedness and Prevention committees and lists of beneficiaries (in rural areas) are drawn up by the peasant associations and confirmed by the community. Government relief agencies usua lly do not actively participate in the selection of beneficiaries, and only have an observer function during distributions.

Dilution of the ration scale results primarily from the political incentive of local authorities in charge of determining relief recipients to cover a larger number of beneficiaries with the available relief food, rather then target more restrictiv ely to the most needy. Also, the mass mobilization targeting policy has created some confusion, as the non-benefiting community work participants do not consider their socio-economic situation very different from that of those receiving aid, and th erefore feel equally entitled to get relief food. Furthermore, if only the poorest of a community are entitled to food assistance, important opportunities for disaster mitigation will be missed, as farmers near the poverty line will be forced to se ll productive assets to compensate for food assistance.

Clarification on targeting criteria, and training in this area for field staff of government and NGO implementing partners and local government officials, is needed. The joint study planned for early 1996 by CDPP, NGOs and the UN will provide addit ional information on targeting practices and should be a source of recommendations on future targeting strategies.

Gender Aspects

Participation of women and girls in EGS work activities is equitable, with some variation according to the nature of the work task, and in some cases has reached 60 percent. However, EGS evaluations have found that women are not adequately represen ted in the elected community structures responsible for choosing EGS activities and selecting food aid beneficiaries; also, no women were employed as project staff and only a few as work gang leaders were women. Future projects should have specific targets for involvement of women in management and participation.

Female headed households constitute the largest proportion of those classified as "not able" to participate in EGS activities, due to lack of able labour. In the Merti Jeju EGS project, only 47 percent of female headed households in the most vulner able group participated, compared to 71 percent of male headed households. This points to a need for improvements in the current system of targeting and continuation of free food safety nets for those unable to work.

Pond projects were found to bring immediate benefits to women by decreasing the time they spent fetching water for cooking and drinking.

1996 Food Aid Planning and Implementation

Food Aid Intervention Strategies

Multi-year pledges: Depending on resources through the annual pledge mechanism is a significant constraint to a more efficient use of relief food aid. The Mission recommends that donors consider making multi-year pledges to allow Government and NG O implementing partners a longer term horizon, especially for timely planning and implementation of EGS activities and setting of longer term objectives. For example, 1996 donor allocations of additional relief food could be used to support 1997 EG S activities.

EFSR: It is recommended that the capacity of the EFSR be increased to 307 000 tons, through additional pledges of around 100 000 tons, to be purchased locally.

EGS: Further clarification is required of the difference between EGS and other means of implementing labour intensive public works, especially with respect to objectives and targeting. For maximum accountability and work quality, EGS programmes sh ould have the following characteristics:

1. labour-intensive activities;

2. a shelf of projects ready in advance of need;

3. serves as employment of last resort;

4. capacity to expand and contract the level of employment offered according to need due to transitory variation in food insecurity;

5. a method of targeting to select the most needy for employment in the project, and the food-deficit areas for activities; and

6. the creation of sustainable assets which help to reduce the future vulnerability of the community to food insecurity.

7. integration with relief and development plans of government line departments.

8. participation is voluntary and participants receive a food allocation based on work performed.

9. planned work satisfies technical criteria and work norms.

10. community participation in planning and selecting activities and choosing beneficiaries.

The Mission recommends that food aid be given on a community work contract basis, whereby the whole community receives food for completing specific activities and selects its most vulnerable members to receive a share on a free relief basis from th e total amount of food provided.

In the future, directly targeted food aid should only be necessary where the beneficiaries lack both economic and physical access to food.

Since the areas affected by severe food shortages in 1996 are chronically food deficit, with limited agricultural potential, food aid can only bring temporary relief until much larger structural adjustments are initiated. Market support, both throu gh improved infrastructure as well as through supply stabilization, will have to become an increasingly important function of food aid.

Improved Assessment and Monitoring: While appropriate for food balance sheet calculations on a national scale, the current agricultural data available at the zonal level is not adequate for vulnerability assessments which aim to identify food-defi cit areas and to quantify their food deficit. The Mission recommends that FAO, working with relevant government ministries, routinely produce dissagregated agricultural production data for future vulnerability assessment purposes.

An integrated institutional framework for vulnerability assessment and early warning surveillance is needed in Ethiopia to improve effectiveness of: 1) targeting emergency resource allocations by identifying areas and socio-economic groups subject to acute food insecurity; and 2) planning appropriate mitigation and development interventions to alleviate chronic vulnerability. The Mission supports efforts to coordinate the activities of food aid actors involved in vulnerability assessment th rough a Steering Committee (FEWS, GIEWS and, NGOs) to develop a common methodology and system for data collection and analysis.

A monitoring and evaluation system for EGS should be standardized and institutionalized to provide information on project inputs, quantity and quality of outputs/achievements, performance, effects on the community and commodity movement for improve d planning and management. The monitoring system, formats and procedures that have been developed and tested for the WFP-supported project 2488 provide a good point of departure. Monitoring and reporting systems for food distribution and EGS should include information on beneficiaries disaggregated by gender.

Mid-year review: The Mission recommends a mid-year review be conducted to: 1) assess the status of efforts to develop an integrated vulnerability assessment institutional framework; 2) review the inclusion of oil in the general ration; and 3) det ermine how best to produce appropriate agricultural data inputs.


During 1996, most of the relief/regular food aid requirements will be resourced locally. This will reduce use of the Eritrean ports and shift trucking operations towards surplus food producing areas. Increased local purchasing will require increase d logistical flexibility as well as improved delivery monitoring systems.

Generally speaking, the national logistics infrastructure is in the process of considerable improvements. The FDRE decided in late 1995 to privatize Government and NGO-owned long-haul relief fleets. However, significant shortfalls in short-haul tra nsport capacity still remain, which the Government hopes will also be taken up by the private sector. Progress towards dismantling the monopoly of the state owned shipping, clearing and forwarding agent, has been disappointing, resulting in a conti nuing high level of costs to relief agencies. The existence of a functioning EFSR contributed significantly to better use of available transport capacities.

Return to Menu