UNITED NATIONS  DEVELOPMENT PROGRAMME Emergencies Unit for Ethiopia

 
I. EXECUTIVE SUMMARY

A combination of favorable main season rains, increased fertilizer sales and improved extension services led to a record harvest of 11.7 million tons in 1996, and predictions that for the first time in recent history Ethiopia would be food self-sufficient. The FAO/WFP Crop Assessment Mission projected surplus production amounting to 492,000 M/T available for local purchase and 400,000 M/T available for cross border trade. Despite this initially very favorable picture, over the past few months there have been signs that food security in many parts of the country remains something that cannot be taken for granted.

Against a backdrop of steadily rising food aid requirements, Food Security Reserve stocks have fallen to a low of some 71,000 tons. Efforts to replenish the reserve through local purchases have faced delays and slower than planned deliveries. With growing concern that a less than favorable main season harvest might coincide with low national food stocks and a poor external pipeline, donors are urged to take urgent measures to ensure government appeals for additional food aid are quickly and adequately met, including the use of imports if needed.

The unsatisfactory short-season rains in the pastoral areas that affected much of Kenya and southern Somalia in late 1996, also had a serious impact on the pastoral economy of south-eastern Ethiopia. Unusual migration into Ethiopia from Somalia and Kenya placed further pressure on already depleted water and grazing leading the Government to revise its original estimate of relief food needs of just under 200,000 tons for 1997 upwards by an additional 90,380 tons. With a poor start to the belg rains, in May the Government was obliged to raise the relief requirement again, this time to an overall figure of 329,450 M/T. Despite a very disappointing donor response to these various appeals, relief operations have so far proceeded relatively well, mainly by falling back on stocks carried-over from 1996 and through heavy use of the Emergency Food Security Reserve. However, replenishment of the reserve through local purchase programme supported by WFP, the European Union and some NGOs has faced delays due to the slower than expected delivery by suppliers.

The period June to September is of critical importance for food production in Ethiopia, with the performance of the main kiremt rains having a direct impact on the main season harvest. Projections so far indicate an erratic and less than favorable 1997 kiremt. Though the rains began on time and to the end of July have been well distributed and sufficient, the occurrence of a significant warm event in the eastern equatorial pacific (the so-called El-Niño event) is expected by experts to result in very erratic rainfall over the Horn of Africa, producing both heavy rainfall and drought in different areas and the likely early withdrawal of the rains.

Due to the poor belg rains, which are also important for land preparation, the area under cultivation for the main season may fall somewhat this year. Adding further to fears that the country might be facing a poor meher harvest are the first sales figures for fertilizer which indicate that usage may be reduced from last year’s record volume of 251,000 M/T. Now that the initial infestation of armyworms is declining following successful control operations and the onset of the main rains, concerns of serious pest damage have diminished. However, there are signs that other pests and plant diseases such as stalk-borer, shoot-fly, coffee berry disease and Wello Bush Cricket, are on the increase and could lead to significant crop losses in some areas, especially in the already food insecure northern highlands. Perhaps reflecting this less than positive outlook, cereal wholesale prices in Addis Ababa are presently more than 20% above that recorded for the same time last year and the volumes being passed through the main markets are reduced.

The relatively poor response to the Government appeals for assistance and the high degree of reliance on locally purchased grain has given rise to a very weak external food pipeline. Between July and the end of the year, scheduled emergency food aid shipments amount to only 72,031 tons, with a significant portion already committed as repayments to the Food Security Reserve.

With these factors in mind it is essential that donors take steps to accelerate the replenishment of the reserve and view sympathetically any additional requests for food aid arising from the poor belg. Furthermore, donors are urged to be flexible when considering whether repayments and new pledges are made through local purchases or imports. Finally, the progress of the main rainy season should be monitored closely over the next two to three months, and appropriate action taken to ensure adequate relief food stocks are available in-country by the end of 1997 should a less than favorable harvest be anticipated.

 
II. WEATHER

Agriculture in Ethiopia is rain dependent and both timing and quantity are of critical importance. In terms of production, the right amount of rain at the right time is far more important than the use of fertilizer, improved seeds, agro-chemicals or irrigation.

The 1997 Belg rains were generally poor but the main Kiremt rainy season started on time in most areas and throughout the month of June and into mid July the rains were generally normal to above normal. However, the long-term weather predictions are not favorable.

El Niño, something that will feature more and more in the international news over the next few months, is an anomalous event of warmer than normal sea surface temperatures in the Central and/or East Equatorial Pacific Ocean. Even though these temperature increases often seem minor they have an enormous affect on weather patterns in many parts of the world. During the April/May/June quarter there was a fairly significant "El Niño event" -i.e. an abnormally high increase in water temperature and "by the summer of 1997, an El Niño event is in progress, and comparisons with past historical El Niño events are being made. The happening in the atmosphere are being actively monitoring by the scientific community, and the expect climate response to this current event continues to be assessed"
 

El Niño effect as of 28/06/97: NOAA Climate Diagnostics Center
 

In terms of Ethiopia, the National Meteorological Services Agency (NMSA), as a consequence of El Niño, in their Long-range Projection for the Belg and Kiremt season issued on 22 May stated that:

The potentially erratic nature of the Kiremt could mean both excessive rainfall and drought. For instance, during the second decade of July (11-20 July) a number of NMSA reporting stations in western and the western parts of central Ethiopia recorded rainfall in excess of 110 mm and some areas of recorded 31 to 57 mm in a single day while areas in North and South Omo, North and South Wello and North Shewa continue to have insufficient rain.

Excessive rainfall, particularly if it comes at the wrong time, can be as damaging to crops as drought. If these long-range projections are correct, pocket areas may loose crops because of abnormally high rainfall while the traditionally vulnerable areas of the country may face drought. Also, as anomalies in the El Niño can affect weather patterns in widely different parts of the world, other countries may also face flooding and/or droughts, placing additional demands of limited relief resources.

III. CROP SITUATION

A. The Agriculture Seasons in Ethiopia

The Belg season or short rains fall between late January and roughly mid-May while the main rains, the Kiremt, begin sometime in June and usually finish at the end of September. The main agriculture season coincides with the Kiremt and is referred to as the Meher and most of Ethiopia’s agriculture production comes from the Meher season. These two production seasons refer to when crops are harvested and Belg production generally refers to crops harvested before the end of August and the Meher to crops harvested after this date, usually from October/November to December or January of the following year. The two main types of Meher crops are the long-cycle crops planted during the Belg rains but not harvested until late in the year and short-cycle Meher crops planted after the main rains begin (June/July) and harvested in November/December. The Belg rains are generally much more unreliable than the Kiremt and it is not unusually for the Belg to be erratic and unevenly distributed. The pastoral areas have a different pattern of rains with the small rains beginning in September/October and the main rains beginning in March/April. Like the Belg, rains in the pastoral areas are often unreliable.

B . The 1997 Belg Rains

Belg cultivation areas cover roughly 9% of the land area and in a good year account for 5 to 7% of the total production; however, in certain Belg dependent areas like parts of North Shewa, North and South Wello, Oromiya and South Tigray zones Belg crops can account for 25% to 50% of total production.

The 1997 Belg was not favorable and can generally be classified as weak and below normal, being late, unevenly distributed and with poor coverage. In addition to planting delays and anticipated reductions in production of planted areas because of moisture stress, the actual area planted in Belg crops was significantly below the original targets.

C. Prospects for the Meher Crop

In addition to the adverse effects a poor Belg has on the food supply in the Belg dependent areas, poor Belg rains can also have a direct impact on the main cropping season, the Meher:

• Land preparation for Meher crops has been delayed in some areas, leaving these crops more susceptible to moisture stress if there is an early withdrawal of the main rains.
• As noted in the table below, long-cycle crops, which are also more drought resistant and have a higher yield per hectare, normally account for about 50% of grain production. (Last year maize and sorghum accounted for 42% of the cereal and pulse production of 11.7 million M/T.) These crops are planted during the Belg season but harvested late in the year. This year, as in previous years with a poor Belg, some of these long-cycle Meher crops will have to be replaced with lower yielding and often less drought resistant short-cycle crops.

• Some of long-cycle crops planted in the Belg season are in poor condition, although they may recover if the present rains continue to be timely and well distributed.

• There is a shortage of improved seeds and in areas where re-planting is necessary additional improved seeds may not be available.

• In some areas where there are normally both Belg and Meher crops, the short-cycle Meher crop may not be planted as the land will still be under Belg production.

• The Belg is important for regenerating pasture areas and draught animals, essential for plowing in the highland areas, may not be in optimum condition, particularly as many highland areas are already facing a shortage of pasture land. SCF(UK), for instance, reported that 83% of the surveyed villages in South Wello and North Shewa (Menz and Gishe) and 81% of the surveyed villages in Southern Tigray reported that livestock condition was poor or very poor. In 1996, on the other hand, "...plentiful and well distributed rains throughout the year not only encouraged planting and supported crop development during the season in most zones but also provided optimal conditions by a draught animal force in excellent condition...."

Other factors that may affect the Meher crop or overall production include:

• There have been anecdotal reports that in some areas farmers, because of low and/or fluctuating grain prices, are switching from staple grains to cash crops or are switching from high yielding but low priced grains such as maize to higher value but lower yielding crops such as teff. In many cases cash crops may be the best alternative but if this trend is widespread there could be a reduction in overall grain production.

• Among the factors noted by last year’s FAO/WFP Crop Assessment Mission as contributing to the increase in production was the 10 fold expansion in the National Extension Programme from 20,000 hectares in 1995 to 200,000 hectares in 1996. This programme, which involves the use of fertilizer, improved seeds, agro-chemicals and extension, will also be expanded in 1997 but the increase will not be nearly as dramatic as the jump from 1995 to 1996. Also, the effectiveness of this programme may be somewhat curtailed by the same factors that may limit this year’s Meher production.

• Two other factors contributing to last year’s excellent production were an estimated 5.6% increase in areas planted in cereals and pulses and an 15% increase in cereal yields. This year, for a variety of reasons, it is possible that not as much land will be planted as last year and yields per hectare may not be as good.

Fertilizer is important to Ethiopian agriculture but in terms of overall production it is not nearly as important as timely, well-distributed rains of the right amount. In the cropping year 1996, which was a surplus year for Ethiopia, total grain and pulse production was estimated by the FAO Crop Assessment Mission to be 11.7 million M/T and fertilizer sales in 1996 were 252,261 M/T. Other things being equal, each kilo of fertilizer means a five-fold increase in production or about 1.2 million M/T of additional production in 1996 because of fertilizer, only about 10 to 11% of the total 1996 grain and pulse production. However, it is important to note that in food deficit years any decrease in fertilizer use means a theoretical increase in the country’s overall food requirement.

Fertilizer sales also appear to be lagging behind the pace of 1996. As of 30 June last year reported sales were 138,427 M/T, 40% of the target and 55% of the final sales figures while as of the end of June this year sales were 79,530 M/T or only 27% of the revised target of 300,000 M/T. Fertilizer sales reporting is usually late but this comparison is nonetheless disturbing. Several theories have been advanced for the slow fertilizer sales in 1997:

• Fertilizer was deregulated in 1997 and for first time there was no fertilizer subsidy and prices increased from roughly birr 190 per quintal (100 kgs) in 1996 to between birr 240 and 250 per quintal, a 30% increase.

• Previously fertilizer prices were fixed throughout the country and farmers knew prices early in the year and could plan accordingly. (In 1993, when there was a delay in announcing fertilizer prices there was a dramatic drop in sales. See Table 3.) Under the new system, although the wholesale price of fertilizer is fixed, transport costs will vary depending on the destination and farmers will not know the actual price until the fertilizer is put on sale in their area.

• Grain prices, particularly maize, remained low throughout 1996 and did not follow the usual trend of increasing around the middle of the year. In 1997 prices picked up again and, until very recently, have followed a more "normal" pattern. However, farmers may still be leery of these wide price fluctuations.

• Normally substantial amounts of fertilizer are used with long-cycle crops planted in the Belg season but harvested in November/December. Because of this year’s poor Belg rains less fertilizer may have been used on the long-cycle crops than was anticipated.

• NIFA believes that the low sales figures as of the end of June also reflect high credit charges and farmers’ reluctance to purchase until the last possible moment.

• Territorial disputes between various fertilizer suppliers and the decentralization of credit and sales responsibilities has caused a certain amount of confusion and more effort may have been spent in defending sales territories than in promoting sales.

• Some farmers have not been able to pay back last year’s fertilizer loans and so are not eligible for new loans this year.

E. Plant Pests and Diseases

Although there have been fairly heavy armyworm infestations reported, local control measures, supported by spraying operations undertaken by the various Agriculture Bureaus, have generally been able to control the situation and with the arrival of the main rains the threat of further infestations has receded. Damage has not been as extensive as in 1994 but some pocket areas will undoubtedly face crop losses.
 

The Ministry of Agriculture also reports a Quela quela bird outbreak in the Somali National Regional State in Dehnan Wereda of Gode Zone with an estimated population of 7.5 million birds. Spraying operations were carried out in late July with an estimated success rate of 97%. Surveys are also underway in the southern Rift Valley.

Although it would appear that migratory pests will not be a major problem this year, local, pests and diseases (stalk borer in maize and sorghum, shoot fly in teff and barely, stem rust in wheat and barley, potato blight, enset bacterial wilt, coffee berry disease, termites and Wello bush cricket, wild animals and rodents), although they receive less publicity, traditionally cause much more damage than migratory pests such as armyworm, locusts or Quela quela birds and pre- and post-harvest losses remain very significant at between 12 and 20%. Also, unlike migratory pests, there is no government subsidy for the control of local pests and diseases unless there is a major outbreak.
 

IV. THE 1997 APPEALS AND RELIEF OPERATIONS

A. 1997 DPPC APPEALS

The 1996 crop year was exceptionally good representing "...an increase of some 20 percent over the MoA’s final estimates for last year’s (1995) Meher harvest...." There were however pockets in many of the traditionally vulnerable areas that required relief food assistance and the Disaster Prevention and Preparedness Commission (DPPC) in their December 1996 Appeal requested a total of 199,846 M/T of emergency relief food. The pastoral areas, which have a different weather pattern were, as usual, not included in this appeal. Throughout the first quarter of 1997 a number of special assessment missions to the pastoral areas were undertaken because of the very poor small rains in these areas and at the end of February the DPPC issued a Special Appeal to Assist Drought Victims in Somalia, Borena and Bale Zones of Oromiya and South Omo Zone of SNNPRS Regions covering an additional 90,380 M/T. In May 1997 a DPPC Early Warning System report raised the total food requirement to 313,673 while the recent DPPC Relief Plan of Operation (June 1997) increased the figure for all areas to 329,450 M/T.

Donor response to these appeals has generally been weak and even excluding additional food that may be required because of Belg losses there is still a shortfall in pledges of about 88,000 M/T.
 

B. "Regular" and "Emergency" Food Aid

The issue of "Regular" and "Emergency" food aid is often confusing, particularly as at the field level both types of programmes often exist side by side and there is little real distinction between Food for Work (a "Regular" programme) and Employment Generation Schemes, (an "Emergency" intervention). As used in this paper "Regular" programmes refer to long-term development type of food aid programmes and includes Food for Work (FFW) and the urban programmes and school feeding programmes of WFP.

Most "Regular" programmes are now implemented in vulnerable areas and are designed to alleviate the root causes of food insecurity. In food deficit years where, in addition to an emergency requirement there is also a structural deficit, any food imported or purchased locally for a "Regular" programme also helps meet the country’s food needs and is considered as a pledge against the country’s total food requirement. However, in "surplus" years such as 1996/97 the issue becomes somewhat more complicated as not all "Regular" food aid is programmed in the specific areas that need emergency food assistance nor does the timing of "Regular" food distributions always correspond to the times of greatest emergency food need. The distinction between "Regular" food aid and "Emergency" food has some bearing on food availability for emergency programmes and remains a point of discussion between the donor community and government.

C. Response to the 1997 DPPC Appeals

Generally donor response to the December DPPC appeal was poor and there has been only limited response to subsequent appeals issued at the end of February, in May and then through the June Relief Plan of Operations.
 

Notes: a. Not included in this list of pledges is a notional pledge from the USA of an additional pledge of 20,000 M/T for the DPPC under Title II (emergency relief).

b. This list of pledges includes responses to the relief requirements only and does not include refugee/returnee food pledges nor 24,000 M/T pledge by the EU to the Emergency Food Security Reserve to bring the reserve up to the full target figure of 307,000 M/T.

Even excluding additional food that may be required because of the Belg losses, the gap between pledges and appeals still stands at a formidable 89,000 M/T.

D. Relief Operations

The failure late last year of the small rains in the pastoral areas put most of the Somalia National Regional State (region 5), Borena and Bale Zones of Oromiya and South Omo Zone of SNNPRS region under severe strain with limited pasture available for the herds and water for both humans and livestock becoming more and more critical. This drought also affected other pastoral areas in the region and the influx of herds and people from Kenya and Somalia compounded the problem. Although the main rains in the pastoral areas came on time in very early April and alleviated the most acute water problems, herds in many areas were decimated and the DPPC was forced to mount a major relief operation in these areas. In order to meet this unanticipated demand the DPPC, at a time when pledges to the December appeal remained extremely weak, "borrowed" food from highland allocations with the hopes that it could eventually be repaid through new pledges. Although there were some additional pledges from several donors, the overall pipeline (both local purchase and imports) remained weak and many highland areas of the country did not have enough food resources to cover both their immediate requirements and the additional food needed to pre-position emergency supplies before areas were cut off by the main rains.

The other major area of concern was Welayita, a traditionally food deficit area that also suffered because of a delay in the Belg rains. Allocations to Welayita had to be increased significantly in order to avoid a crisis such as the area faced in 1994, putting a further strain on limited resources.

Although a major crisis was avoided, two areas remain of concern:

• In cases where relief distributions are delayed beneficiaries may be forced to sell assets in or migrate to towns order to provide for their families. Any loss of assets means that recovery/rehabilitation will be take longer and be more expensive.

• Areas where pre-positioning was not possible may emerge at the end of the rainy season in urgent need of food assistance and supplies may have to be moved to these areas as soon as the roads re-open.

A. Cereal Price Trends

Cereal prices traditionally increase during the "lean" period between May/June and October/November but over the last two months wholesale grain prices have increased significantly:
 

Although grain price increases are normal for this time of year and have now stabilized somewhat, other factors that may have influenced this trend include an increased demand for seeds, an anticipated poor Belg harvest and speculation. In the case of white maize, the Grain Market Research Project speculates that another reason for the sharp price increase may be that the maize grown last year is reported to be highly susceptible to termites and other pests during storage, thus forcing farmers to sell their excess production early in the season.

Nominal wholesale prices in Addis Ababa of the major grains are, on the average, 22% higher than June a year ago with white maize recording the highest price jump of 38.5%. Again, the poor Belg, speculation and "... the late Meher rain could possibly be the reasons for the higher prices observed this year."

B. The Emergency Food Security Reserve (EFSR)

The EFSR was established in 1992 with an initial stock of 52,442 M/T. Major donors to the reserve include the European Union, WFP, the Netherlands, CIDA, Australia and the Ethiopian Government and additional pledges over the last few years have brought the reserves holdings up to 267,247 M/T with the balance of 39,753 M/T needed to bring the Reserve up to the target of 307,000 M/T already pledged but not scheduled for delivery until very late in 1997 or in 1998. The Reserve was established to bridge the gap between pledges and the actual availability of the food (often six months) and since its establishment the Reserve has played an increasingly important role in meeting immediate food needs through the release of food stocks against confirmed pledges (See Table 8). The Reserve was specifically set up as a revolving fund with loans only being made against confirmed pledges and grants from the Reserve are only made in times of extreme emergencies. Normally the minimum level set for the Reserve is 25% of its holdings, currently about 66,800 M/T (25% of 267,247 M/T in physical stock and outstanding loans) but because of "....seriousness of the situation, the Board had decided this minimum level of the stock to be utilized very cautiously and to be utilized only for loans requested for emergency relief interventions."

The utility of the Reserve is directly related to the amount of physical stocks it has available for fresh loans - i.e. for the Reserve to continue to function effectively and meet its original objective as a bridging mechanism between pledges and availability it is vitally important that loans be repaid quickly. According the EFSR Administration as of 29 July the position of the Reserve was as follows:
 

The repayment schedule for balance of outstanding loans of 106,212 M/T is not fixed nor is it clear whether these remaining repayments will be made through local purchases or through imports.

Additional repayments of the outstanding 106,212 M/T are important as it is very likely that any new pledges that may be made against either the current DPPC appeals or any appeal for the Belg will be used to borrow food from EFSR stocks, further depleting the amount of food available in early 1998. The urgency of the repayment problem is also covered in the EFSRA’s most recent Newsletter where the Board:

Over the last few years carryover stocks have played a significant role in meeting food needs in the early part of the new year. In a sense carryover stocks are like the EFSR but have the added advantage of not requiring new pledges.

Carryover stocks include both physical stocks in-country at the end of the calendar year and previous year’s pledges due to arrive in the following year. The currently agreed upon carryover figure by WFP and the DPPC from 1996 into 1997 is 155,698 M/T, composed of 68,726 M/T in carryover pledges and 88,267 M/T of physical stock in-country as of 31 December 1996.

Although Table 9 lists over 67,000 M/T as carryover pledges, this figure must be used with some caution when trying to determine exactly how much food might really be available to meet the 1997 emergency food needs:

• The 10,705 M/T EC carryover is the balance due against the 1996 pledge of 75,000 M/T to the Emergency Food Security Reserve. When delivered to the EFSR this 10,705 will be available for loans against new pledges.

• The USAID 400 M/T is oil for monetization and its arrival will not affect the 1997 emergency food requirements.

• 19,900 M/T of the WFP carryover of 48,527 is for Regular Programmes and only 28,100 is emergency carryover. If the areas to be covered by this 19,900 M/T Regular Food Aid correspond to the locales specifically in need of emergency assistance then this food will help meet the emergency food requirement. (Also see section IV. B. on Regular and Emergency food aid.)

• Both the Canadian 3,380 M/T and the German 3,000 M/T are also carryovers of Regular Food Aid and, like the WFP 19,900M/T, they will only help meet the 1997 emergency food aid requirements if they are utilized in areas in need of emergency assistance in 1997.

Although WFP has been purchasing pulses locally for some years, significant local purchases of cereals did not begin until 1996 when programmes were launched by the European Union, WFP and EuronAid. The agriculture year of 1996 was even better than 1995 and with the FAO/WFP Crop Assessment Mission projecting "...a surplus of 492,000 tons available for donor supported local food aid purchases and triangular transactions" the local purchase programme was expanded in 1997.

Currently the local purchase situation is somewhat confusing as there are local purchase programmes from 1995 and 1996 carried over into 1997 as well as 1997 programmes covering repayments to the Emergency Food Security Reserve, new pledges to the EFSR, local purchases for regular Food for Work programmes, for emergency relief programmes, for refugee programmes and for export. A number of NGOs and bilaterals are also involved in local purchase programmes, the government has done some local purchasing for export and, in addition to the three main cereals (maize, wheat and sorghum), pulses as well as "Famix", "Fenamix", "Faffa", salt and biscuits have also been or are scheduled to be purchased locally.

As noted in the above table, the European Commission and EuronAid has suspended their local purchase operations because of the large number of suppliers who have defaulted on contracts. WFP, which is by far the largest purchaser this year, however, reports that presently their local purchase programme is going well with deliveries of between 2,000 and 4,000 M/T per week and they still hope to have a total roughly 50,000 to 60,000 M/T delivered as the end of August and 75,000 M/T as of the October/November.

Although WFP had some contractors who defaulted on contracts early in the year, once WFP began to verify stocks before signing contracts defaulting decreased. Also, the majority of WFP’s pending 1997 contract will be through the Ethiopian Grain Trading Enterprise (EGTE), enabling WFP to be assured that quantities will be delivered on time. Direct purchases from EGTE have also speeded up the process.

A number of different theories have been advanced to try and explain the slow pace of this year’s local purchase programme:

• The suppliers contented that the decision by the main agencies involved in local purchases not to provide advances has been a significant impediment to the successful implementation of contracts. The purchasing agencies, however, feel that advances were unnecessary and an administrative headache, particularly when dealing with defaulters. Instead of pre-payments, the size of individual tenders was reduced from 3,000 M/T down to 500 M/T and payments, if necessary, were made on the basis of quantities delivered. WFP will also assist small merchants with the supply of bags or the provision of transport.

• Some agencies and donors feel that the 1996 Meher production was over estimated and that there was not as much excess production as originally anticipated. It is difficult to gauge whether this is correct or not but another explanation might be that the overall production estimates were reasonably accurate but that much of the excess production is at the farm level in the form of one or two quintals. Individual farmers may have found it either difficult or uneconomical to move grain to the local markets or these small quantities are simply not accessible to merchants operating on a very limited overhead.

• Rural Ethiopia remains extremely isolated with a very weak rural road network and only a fledgling marketing system. Moving excess production to markets accessible to the bigger merchants is still difficult and in some cases uneconomical, particularly in the early part of the year when grain prices are often depressed.

• Recently prices have increased to such an extent that suppliers are unable to deliver without incurring a loss and prefer to forfeit the security deposit rather than fulfilling an uneconomical contract.

• Earlier in the year suppliers also complained that the quality specifications were too stringent and they lacked the facilities to clean grain prior to delivery. The EFSR has, however, shown greater flexibility in quality control and this should no longer be a major impediment.

• Some suppliers were overly optimistic on what they could deliver and when they could deliver it and, as a result, had successful bids pending with several agencies at the same time.

• The FAO/WFP December Crop and Food Assessment Mission anticipated "...a large increase in cross border exports in 1997, to a total of about 400,000 tons." Informal trade, particularly to Eritrea, may have been higher than anticipated.

At this time the direct impact of the local purchase programmes on 1997 emergency food needs is rather limited as most stocks currently being purchased will be used to repay the Emergency Food Security Reserve; however, any delays in the local purchase programme could have a very negative effect on food available from the EFSR for further loans:

• Although 39,753 M/T in pledges exist to bring the Reserve up to the target of 307,000 M/T, under present plans this food will not be purchased until late 1997 on sometime in 1998 and total holdings still stand at 267,247 M/T.

• Physical stocks in the Emergency Food Security Reserve remain extremely low because of delays in local purchases in the early part of the year. This means that even if new pledges are made there may not be sufficient stocks in the Reserve to make significant new loans.

• Cereal wholesale prices have increased significantly since the beginning of the year (the price of maize almost doubled from 30 to 70 birr a quintal in December/January to birr 67 to 116 per quintal at the end of June) and, although they seem to have stabilized during the first two weeks of July, they are likely to remain fairly high until the new harvest begins to come on the market in November. In cases where donors have a fixed amount of money available the amount of grain they can now purchase may be less than they originally anticipated or they will have to pay more for the same amount of grain.

• If grain prices remain high or only limited stocks are available until the new harvest then some repayments to the Reserve that were scheduled to be made in 1997 may have to be postponed until 1998.
• Grain merchants have already complained that there are now limited stocks in the market available for purchase. Information collected from national household surveys in 1996 indicate that by May around 90% of farmer grain sales have already taken place. If this is the case in 1997 then, aside from stocks held by EGTE, there will be limited opportunities for additional local purchases until the new harvest comes on the market.

• If the 1997 Meher harvest is limited then large quantities of grain may be available for local purchase in 1998 as was originally anticipated.

E. Port Stocks

Assab port stocks as of 18 July were 7,350 M/T while Djibouti stocks were 6,375 M/T; however, the Assab stocks are entirely oil scheduled for monitization while the Djibouti stocks are for the refugee programmes and there are no relief stocks in the port at the moment.

F. In-Country Stocks

Up-to-date figures on in-country stock balances of emergency food supplies are not available but given the limited amount of food available and the high demands for both immediate relief distributions and for pre-positioning it is unlikely that in-country stocks are significant.

G. Shipment Schedule (Pipeline)

In addition to a weak response from the donors to the DPPC’s various appeals, many of the pledges that have been made are to be met through local purchases. Consequently, the pipeline is particularly weak and even including carryover stocks from 1996 into 1997 that are still due (see Table 9), the total amount of imported emergency relief assistance scheduled between mid July and the end of the year is only 72,031 M/T. Even more important, most of the emergency relief imports will be used to repay the Emergency Food Security Reserve for loans already taken and, in most cases, already distributed. Although EFSR loan repayments are essential, this also means that this food will not be available for distribution unless there are additional pledges.

VI. SUMMARY AND RECOMMENDATIONS

Although relief food is in short supply in some areas, presently Ethiopia does not face a major crisis nor is the situation critical at the moment; however, the poor Belg, inability to pre-position sufficient relief supplies in many vulnerable areas, the possibility of a more "normal" Meher harvest, the potential for low carryover stocks into 1998, the depleted EFSR, limited pledges against the appeals and a poor pipeline could mean that Ethiopia will enter 1998 with very limited relief food reserves but relatively high relief food needs.

• The progress of the main rainy season should be monitored closely over the next two to three months, particularly as most long-term forecasts predict erratic rains (both too much and too little) and the possibility of an early withdrawal. Also requiring close monitoring will be those areas where it was not possible to pre-position relief supplies prior to the main rains.

• Donors should review the current DPPC requests for food assistance and view sympathetically any additional requests for food aid that may be needed because of the poor Belg.

• Prompt repayments to the Emergency Food Security Reserve are important as are additional pledges to meet the current shortfall or any additional appeal that may be issued by the DPPC because of the poor Belg. In view of the uncertainty regarding the availability of grain for local purchase, donors should be flexible in terms of whether repayments and new pledges are made through local purchases or imports and heed the recent DPPC request for additional imports.

• Donors supporting both food for development type programmes (FFW) as well as relief programmes should, in critical situations, be flexible in switching resources from one programme to another.

• The two recent studies done on food aid targeting have both shown that the most effective use of relief food aid is obtained by ensuring the right areas are selected rather than attempting to concentrate on household selection and the DPPC should continue to give high priority to refining and improving the targeting procedures and systems.

• Discussions should continue on refining and clarifying the distinctions that may or may not exist between regular food aid and emergency food aid during both deficit years and surplus years.

• Most regular food aid programmes and funded on a multi-year basis, thus facilitating continuity and a smoother flow of resources. Donors should also consider the possibility of developing flexible, multi-year pledging systems for emergencies.

• As exports to Eritrea may have had a significant impact on the availability of grain for local purchase, the current EC study on cereal availability should also try and ascertain the extent of cross border trade to Eritrea, Sudan and Kenya.

• Market/grain availability studies such as the current study being undertaken by the European Commission in cooperation with WFP are important in determining realistic local purchase targets but in order to take advantage of the most favorable prices the next study should take place very early in the new year.

The designations employed and the presentation of material in this document do not imply the expression of any opinion whatsoever of the UN concerning the legal status of any country, territory, city or area of its authorities, or concerning the delimitation of its frontiers or boundaries.

1 August 1997