ETHIOPIA

ETHIOPIA Food Security Update October 2006

Although on-going humanitarian interventions and the good prospects for the meher or main season continue to have a positive impact on the food security of the population, 10.4 million people require humanitarian assistance through the Productive Safety Net Program (7.3 million people) and the emergency program (3.1 million people). This level of food insecurity is expected to persist, especially in pastoral areas. The chronically food insecure areas of eastern Ethiopia and the southern pastoral zones of Oromiya and Somali Regions continue to be the most food insecure areas (Figure 1).

Recent pledges have improved the food aid pipeline through October but significant pledges of food aid are still needed to prevent a pipeline break in November and December. In addition, 60,000 MT of food aid is needed to ensure that there are sufficient stocks to meet some of needs during the first part of 2007, when requirements in pastoral areas tend to peak. Given the current poor performance of the rains in Somali Region, ensuring a healthy pipeline in early 2007 should be a priority. Finally, various sources indicate that the provision of adequate non-food resources continues to be a concern - in 2006, only 37 percent of the total revised non-food requirements of US$ 111 million have been pledged so far. Late and inadequate responses are commonly seen in all non-food sectors, hampering the overall livelihood recovery process, especially in pastoral areas.

In 2006, the PNSP continues to transfer both food and cash to about 7.3 million chronically food insecure people, a dramatic increase from the 4.8 million beneficiaries covered in 2005. The PSNP did not cover the pastoral regions of Somali and Afar in 2005, but covered Afar in 2006. It is not yet clear if the PSNP will cover Somali region in 2007. If the PSNP is not operational in Somali during 2007, the chronically food insecure caseload would have to be covered through emergency food programs.

Figure 1: Percentage of population needing humanitarian assistance in 2006 (Emergency + PSNP)

Data source: Disaster Prevention and Preparedness Agency (DPPA) and Food Security Coordination Bureau (FSCB). Graphics by FEWS NET, September 2006.

• Complete or partial deyr/gu season failures in southern and southeastern Ethiopia affected the lives of millions of pastoralists, limiting or depleting the availability of water, browse and fodder.
• Floods and hailstorms caused the deaths of more than 600 people, with many more unaccounted for, and caused serious damage to crops and household assets during the current meher rainy season (June to September).
• Cereal prices remain at record high levels, showing continuous increases since January 2006; terms-of-trade continue to be unfavorable for pastoralists.
• Tribal and resource-based conflicts occurred between May and July period in southern Oromiya and Somali regions.
• Formal and informal commercial cross-border trade is restricted, due to insecurity and official border closure.

While most of the country is entering the dry season and harvest period, the short-season rains (deyr) normally fall over the livestock-dependent southern and southeastern lowlands of the country between October and December.

This year, these rains are critical to the livelihoods of pastoralists who have suffered two consecutive bad seasons which have resulted in significant reductions in herd sizes and eroding the primary basis of pastoral livelihoods. Normal rains will end the peak hunger season by recharging water sources and replenishing pastures which will sustain livestock through the dry-season from December to March - April. On the other hand, poor rains will prolong and worsen the hunger season and drive these regions back into a severe crisis similar to that seen at the beginning of 2005.

So far the deyr season has not established itself adequately to support a significant improvement in the situation. Light showers and some concentrated rains (which caused flooding) have occurred in parts of Fik and Hamaro districts of Fik Zone, in Dagahbur District of Dagahbur Zone and in Doloado District of Liban and in Gode town. These rains are expected to temper the prevailing crisis of food insecurity by reducing water shortages in the area, but will not support sustained improvements in food security conditions unless they continue. This is evidenced by numerous reports of poor water availability in all southern zones of the region.

Unless rainfall improves during the remainder of the deyr season (especially in November), access to pasture and water will be insufficient to sustain livestock herds and human needs through the dry season until the long-rains begin in March or April 2007. In view of the severity of the food security situation, if the rains are poor the region will descend into another severe crisis unless the necessary preparation measures are taken. Food and non-food interventions must continue in these areas and humanitarian actors should be ready to scale up interventions if needed.

The DPPA-led annual national crop production and need assessment for 2007 for crop dependent areas will start on November 11th, while the pastoral areas assessment covering Somali region and Borena zone of Oromiya region will start on December 2nd and will continue to the 20th of December. The assessments will include all major donors, UN agencies and international NGOs as well as government national early warning (EW) staff, including the newly established Livelihood Integration Unit (LIU).

The household economy approach (HEA) used last year in SNNPR will again be used in the region. This method takes into account sources of food, income, expenditure and coping abilities of households determining if households will face a food gap and if so how big this gap will be. It is envisaged that this method, which will be a bases for monitoring, analyzing and projecting food gap based on livelihood analysis at a household level, will be expanded to Tigray, Amhara, and pastoral areas of Afar and Somali regions next year. Livelihood baselines are being compiled or updated in these regions to allow this to happen.

In addition, a flood impact assessment in the areas affected by the July-September floods has started on October 15th. This assessment will also evaluate the response to date and gaps in relief and rehabilitation needs of the affected population.

The assessment results of this and other non-food needs assessment will form the basis for relief programming for the first six months of 2007 and is due to be released in early January 2007.

Ethiopia, with recurrent droughts has received much more food aid than any other food aid recipient country in the world for the past decade. Ethiopia received an average of 715,000 MT of food aid per year between 1992 and 2003 with a maximum close to 2 million MT in 2003, the worst crisis year in recent times. Food aid imports constitute 25 to 30 percent of the local marketable surplus. This proportion is much higher for wheat as more than 80 percent of food assistance is wheat.

As a result of Ethiopia's chronically high levels of food insecurity, the Government and donors have reoriented their approach focusing on the underlying causes of food insecurity. Thus, the Productive Safety Net Program or PSNP with multiyear commitments from donors has been established with the dual objectives of bridging the consumption gap of chronically food insecure households and engaging them in community-based asset building activities effectively started in March 2005. The PSNP still works through food or cash resource transfers to meet basic needs via labor-based public works or direct support programs.

Despite these developments, the magnitude of food aid being transferred to beneficiaries has not diminished in recent years, nor has there been a significant shift in terms of response modalities. The growing numbers of transitory and chronically food insecure people and increasing volumes of international assistance to meet emergency appeals and annual structural food deficits suggest that at least in the foreseeable future, food aid must continue to be a component of an appropriate response, both in times of surplus and deficit.

Various studies have indicated that humanitarian assistance, particularly food aid has played a significant role in saving lives and protecting livelihoods of the poor in the country. Findings also indicate that food aid recipient households often sell part or even all of their received food aid grains, particularly wheat on the market in order to pay for household expenses or to buy other food items they prefer more for consumption. Surveys further indicate that the quantities of food aid that enter the market vary, may go as high as 30 percent of the marketed volume, depending on the local situation and timing of the food aid. Poor and vulnerable households also release more than 90 percent of their produce before April, which is also the time when prices start to increase in most grain markets. These basic facts are critical in understanding the timing of food aid distribution and its effect on the household economy and local marketing.

Figure 2: Selected Zones for the household survey

Source: EC, Study on Impacts of Food Aid on Grain Market in Ethiopia, August 2006.

Better understanding of the effects of food aid on the domestic grain market and on the household economy, from the perspective of smallholder producers, recipients of food aid, and traders and cooperatives is therefore needed. However, until recently there had not been an in-depth survey conducted to analyze the impacts of food aid on markets. The European Commission (EC) conducted such a study between April and July 2006.

The study covering 684 food aid recipient and non- food aid recipient households and more than 71 traders in the country highlighted the negative effects of food aid on grain traders (and/or suppliers to the market) by decreasing the demand, decreasing interregional grain trade flows, and also causing price and market instability in surplus producing areas (though temporarily). The impact differs depending on the type of grain that is supplied to the market - in some cases households sell high value crops rather than the food aid, which usually brings lower prices than the similar locally produced items. In other cases, households sell wheat at a relatively high price and buy maize at lower prices.

The study also found that lower priced food aid usually decreases prices of a close substitute crop in the market. The negative effects of food aid on grain prices, demand and production can only be reduced if the proportion of imported grains is decreased and the balance is further shifted towards local procurement. It was further concluded that cash aid showed a temporary price hike effect on the market due to increased demand for the major consumable items. This has also been observed to increase market flow between adjacent local markets. This is positive for the producers as it further encourages more production but negative for local net consumers/buyers/recipients of cash aid who can afford less food at higher prices.

The study, which covers 6 food aid recipient and 6 non-food aid recipient districts (woredas), generally supports the assumption that local purchase of food aid and the cash component of the PSNP are more likely to have positive liquidity effects on the market than the distribution of imported food aid. According to this study, imported food aid should complement local purchase rather than the other way around.

In Ethiopia, structural food deficits often occur during periods of localized surplus production. The fact that food deficits are often related to poverty and insufficient market demand provides the rationale for the local purchase of food from surplus areas for redistribution as food aid to deficit areas. Hence, the import of food aid, especially during surplus years, may have localized disincentive effects, which may work to the detriment of small scale peasant farmers who rely on the income from their crop sales to meet their families' needs. Local purchase was found to be a time saving and cost-effective way to procure food aid and also supports domestic production and market development. Local purchase of food aid and provision of cash as aid with other innovative price support approaches such as supporting farmers' associations/cooperatives and facilitating guaranteed credit to traders can help reduce some of the unintended negative effects of food aid on the market.

In general, all kinds of interventions have a direct or an indirect (temporary or long lasting) impact on the market. However, the cereal markets in the country remain weak and thin. Market infrastructure continues to be a serious constraint. It is, therefore, necessary to improve the efficiency and equity of cereal markets. This will help to avoid artificial surpluses and deficits in the grain marketing system. More efficient and equitable markets require investments in transport and storage infrastructure, in processing industries, market price information systems (to minimize information asymmetry), and institutional innovation and policy reforms (especially in the area of border closure) at all levels of the supply chain. An improved financial system is also needed with the capacity to provide credit. It also requires the elimination of government and private systems that limit competition among grain traders and opening markets with neighboring countries to overcome market stagnation due to market thinness. These improvements will help to exploit the potential spatial arbitrage opportunities by moving grain from surplus regions to deficit regions or from rural areas to urban areas. Finally, all interventions should be guided by a serious and robust analysis of marketing cycles, volatilities, chains and price trends.

* This section is a summary of the recent study done by the Delegation of the European Commission to Ethiopia - 'Study on Impacts of Food Aid on Grain Market in Ethiopia, August 2006'.

Cereal prices after reaching their peak level during the lean months between June and August normally decline during the period between September and November as the flow of harvested crops increase into the market. Cereals account for about 90 percent of total food grain production in the country.

This year, during July and August, cereal prices followed their normal seasonal rise, as the marketable surplus of crops declined before the harvest of crops. With some signs of harvest during late September in lowland areas, prices have started to show some signs of stability at about their August levels for the first time since December 2005. Central Statistical Authority sources have also indicated that the consumer price index for cereals in September remained stable at a higher level. However, current overall cereal prices are still high when compared to the same period last year and their respective averages. (Figures 3 and 4).

With the on-going meher harvest, maize prices started to stabilize in September at their extremely high July and August levels of 175 birr/quintal (1 quintal=100lkg) in Addis Ababa and 225 birr/quintal in Jijiga. On the other hand, teff prices in surplus areas (like BahirDar) continued to show an increase despite signs of a good meher harvest.

Currently stable prices are attributed to recently harvested crops (maize, potato, chick peas, etc.) coming onto the market and to grain speculators releasing grains in anticipation of price declines following the harvest. Despite favorable harvest prospects, cereal prices are not yet expected to decline for the coming months and are anticipated to remain at much above average levels.

Price stability at the current record-high levels may help producers get better prices at the start of the harvest season. However, these high prices will have negative repercussions for households who rely on the market for purchases of food (particularly the most vulnerable and poor urban consumers) by sharply eroding their real purchasing power and ability to cope. Without an offsetting increase in other sources of income, persistently high cereal price will result in high levels of food insecurity contributing to increased malnutrition and morbidity. Hence, the Government and donors must regularly monitor prices in food markets (particularly cereal and livestock markets) to quickly identify unusual price behavior. Understanding the causes and consequences of price anomalies will also help decision makers take the appropriate measures.

Should cereal prices decrease in the coming months with the possible increase in supply, there is a high potential that this could result in an improvement in food security in many areas, where the poor dominantly depends on purchases for their staple food consumption. Such decreases in cereal prices, if paralleled by increase or high level stability in livestock prices, would also favor pastoralists through improved terms of trade (which is reported to be still in favor of grain suppliers).

Figure 3: Addis Ababa wholesale prices for selected staple cereals

Figure 4: Real retail prices of white maize in Addis Ababa: 13-month trend and 2001-2005 monthly average

Source: Data archives of FEWS NET/Ethiopia, and Ethiopian Grain Trade Enterprise (EGTE). Graphics by FEWS NET/Ethiopia.
Notes: (1) Prices are expressed in Ethiopian Birr per Quintal. (2) One Quintal = 100 kg; 1 Ethiopian Birr ? 12 US Cents. (3) Average prices (2001-2005) are deflated by their respective consumer price index values.

The main growing season rains in Ethiopia start to wind down in September. The Inter-Tropical Convergence Zone (ITCZ*) starts to gradually retreat from the northeast to southwest, until it moves beyond Ethiopia's borders in late November.

In late September and the beginning of October through November, the ITCZ is normally positioned over the southern lowlands of the country where it produces the short season rains, locally known as deyr/hagaya. Rains usually start in South Omo and Borena zones in the middle of September and progressively move eastward, reaching the Somali Region in October. In a normal year, these rains continue to the end of November. These rains are extremely important for replenishing pasture and water in the pastoral areas of South Omo Zone, SNNPR, Borena Zone, Oromiya Region, and all of the southern zones of Somali Region.

Figure 5 depicts satellite estimates of rainfall distribution during September and early October compared to normal. The rainfall distribution during September was normal to above normal for most meher rain dependent parts of the country except most parts of Afar and eastern Amhara and Tigray. Below normal rains were also received in the pastoral south, particularly in Borena and South Omo zones.

The short season rains started in Borena Zone and southern zones of Somali Region during the third dekad of September and continued in the first dekad of October. However, their amount and distribution has been inadequate so far. Delays of rains have also been observed in South Omo of SNNPR and some parts of Borena Zone of Oromiya Region. Floods have also been reports in Gode zone due to 1 to 2 days of heavy rainfall.

If National Meteorological Agency's (NMA) forecast of normal to above normal rains holds, and if rains continue up to November, the rains should relieve drought stress and improve access to pasture and browse. However, in the worst-affected areas, it may take several consecutive good seasons for herd sizes to increase. Finally, in contrast if the current pattern of rainfall continues in southern zones of Somali Region, it will precipitate another major humanitarian crisis.

In the major crop dependent areas of Ethiopia, the crop harvest started in late September and early October in the northern parts of the country where rains cease earlier. Harvesting is expected to steadily shift to the central and southern highlands of the country and typically peaks in November but continues through December and January in some highland areas of Oromiya Region. During these months, late rains are not desirable and if rains continue beyond the first dekad (10-day period) of October, they will disrupt harvest activities; lower quality and decrease yields, and increasing in-harvest loss of production as fragile cereal grains, especially teff.

Figure 5: Satellite estimated rainfall compared to long-term average (last dekad of July through first dekad of September 2006) and Climate Outlook for October - December 2006.

Source: NOAA Rainfall Image, produced By: USGS-EROS Data Center. The outlook is based on NMA forecast, September 2006. Graphics by FEWS NET.
Note: The percentages in the climate outlook indicate probability of getting seasonal rains with respect to their respective long run averages.

* The Intertropical Convergence Zone (ITCZ) is an area of low pressure that forms where the Northeast Trade Winds meet the Southeast Trade Winds near the earth's equator. As these winds converge, moist air is forced upward. This causes water vapor to condense, or be "squeezed" out, as the air cools and rises, resulting in a band of heavy precipitation around the globe. An exception to this rule occurs when there is an ENSO event (refer next article of this update), during which the ITCZ is deflected toward unusually warm sea surface temperatures in the tropical Pacific.

According to climate scientists, while climate variability is hard to predict, many climatic variations are part of patterns that are coherent on a large scale. Thus skillful - or reliable - prediction is a possibility under certain conditions such as, when certain patterns of Sea Surface Temperature (SST) exist.

El Niño (warm event or warm episode) is an increase in the SSTs in the central and eastern equatorial Pacific Ocean. It is a coupled air and ocean phenomenon with global weather implications. The two processes, one in the ocean, El Niño, and one in the atmosphere, the Southern Oscillation, interact to form what is called El Niño/Southern Oscillation (ENSO). The Southern Oscillation Index (SOI) is a commonly used indicator of the state of the ENSO. Large negative values of the index indicate El Niño episodes. On the other hand, La Niña (a cold event) is the name given to cold sea-surface temperatures in the central Pacific.

El Niño episodes typically start around March to May and last from 12 to 18 months. If the SOI is strongly negative by around June or July, then an El Niño has usually commenced and will last into the following year. El Niño has a return period of four to five years. Weather and climate anomalies that are linked to the appearance of an El Niño event are called ‘tele-connections' or connections over great distances.

According to various studies, climatic variations in Ethiopia are linked to these global phenomena. The tele-connection between El Niño and the Ethiopian seasons has been established by the National Meteorological Agency (NMA). Various droughts in Ethiopia have been linked to the physical processes related to El Niño tele-connections. Empirical observations also show that the rainfall patterns (and also droughts) in Ethiopia are linked to observable changes in SSTs.

The following broad relationships are believed to occur between the belg (February to May) and kiremt (June to September) seasonal rainfalls in Ethiopia and the SST anomalies.

• La Niña (cold event) is strongly associated with deficit belg rainfall.
• La Niña is associated with heavy kiremt rainfall.
• El Niño is associated with normal and above normal belg rainfall (boosts small rainfall season).
• El Niño is associated with decreased and disrupted kiremt rainfall (main season) or widespread meteorological drought or below normal rainfall and wet October to December period.
• El Niño and La Niña may not be the only cause of drought in Ethiopia.

Although not every area is affected severely in each episode, El Niño does lead to an increased likelihood of substantial climate anomalies in Ethiopia. It tends to amplify the climate variability, impose a specific temporal pattern on droughts and heavy rainfall periods, and allows some predictability of these variations. This effect is strongest over lowland and low annual rainfall receiving areas, so is especially relevant to the semi-arid areas.

During El Niño parts of Ethiopia are at a higher risk of suffering from severe drought. Western, southern and southwestern parts of the country are usually flooded between October and December. During La Niña (like 1998), the whole western half of the country tends to be flooded between July and September. However, not every weather problem that takes place during an El Niño can be blamed on it. The Somali Region has not experienced abundant rainfall since the El Niño event in late 1997. Thanks to El Niño's episodic event, the Region received abundant rainfall for the first time in many years. Regional and local conditions can also make El Niño's potential impacts worse or can block them from happening. In addition, according to experts, the utility of improved El Niño forecast systems in early warning system will depend on a number of factors, including the accuracy of the forecasts; variations in impacts between El Niño episodes; the way in which the forecasts are used; multiple sources of forecasts and confounding effects of climate change (that is, in the longer time scales).

According to NMA experts, a weak to moderate El Niño has developed. As a result, on September 29th, the NMA forecast a high likelihood of normal to above-normal rainfall for the period September to December for all parts of the country. The forecast also highlighted the warming of SSTs across the subtropical Atlantic Ocean and Mediterranean Sea generating interactive weather-types.

NMA also forecast the possible occurrence of unseasonable rains from October to December 2006. There is, however, a decrease in the possibility of frost-occurrences. NMA has also predicted that the meher rain withdrawal will be normal, with some areas facing a possible late withdrawal. In most crop dependent areas, the period from October to December is normally dry, during which most crops attain full maturity. Late withdrawal and rains between October and December would interfere with the harvest and could damage the already maturing crops.

On the other hand, predicted normal rains in Gambella and Benishangul Gumuz, western Oromiya and western SNNP Regions will not have any negative impact as the areas are accustomed to excess precipitation during the season. Areas in SNNPR will be able to produce short-cycle crops such as teff and pulses with the extended residual moisture. If the forecast of above-normal rainfall holds true for most of the Somali Region as well as Guji, Borena and Bale Zones of Oromiya Region, the food security situation of pastoralists will significantly improve following regeneration of pasture and increased water availability thereby increasing livestock productivity. Crop-dependent parts of South Omo may also produce some maize.

Since localized month-to-month and dekad-to-dekad variations are likely to occur, the forecasting centers advise users to follow the regular forecast updates provided by the National Meteorological Agency.

* This section is extracted from various sources, including researches by Tsegay Wolde-Georgis, IRI references, Internet Journal of African Studies, Bureau of Meteorology Research Centre in Australia, Lamont-Doherty Earth Observatory, Columbia University, etc.