Issue 27 - Article 3

Food insecurity and aid policies in Ethiopia

July 13, 2004
Jerome Frignet, Action Contre la Faim

The famine forecast for 2002–2003 put Ethiopia once again – albeit briefly – on the international front pages. Ethiopian Prime Minister Meles Zenawi’s comparisons with the situation in 1984–85 were duly relayed by the media, yet the circumstances of these two crises are so different that a repetition now seems almost inconceivable. In the event, the 2002–2003 food crisis did not have the feared lethal intensity, though it was still exceptional in terms of numbers involved: at the end of 2002, the UN spoke of perhaps 15 million people ‘threatened’ by famine. All the attention and resources of the international community were directed towards emergency response, mainly food distribution.

Should we simply rejoice that, 20 years on, early warning and response systems have progressed so far that a large-scale famine is no longer a major risk in a country that so neatly symbolises hunger in Western eyes? Or should we also ask questions about the persistence of such immense food insecurity in a relatively peaceful country? It seems that the attention paid to early warning and emergency response mechanisms, vital as they are, may have diverted minds away from more fundamental issues about the persistence of food insecurity and how to deal with it. The bulk of aid focuses on short-term responses; the deeper causes of food insecurity are largely ignored due to lack of political will on the part of the government, and the essentially reactive policy of donors.

‘Model’ warning and emergency response mechanisms?

Since the ‘hidden famine’ of 1973–74, the Ethiopian government has tried to develop an early warning and emergency response system that would lead the way in sub-Saharan Africa. The system’s growing openness and transparency since the fall of the Mengistu regime has ensured much better support from international donors. In 2003, pledges exceeded 90% of the 1.7m tonnes requested. About 50% of distributions were arranged by the government. The government also has emergency stocks of around 200,000 tons, allowing it to make up for the traditional tardiness of international contributions. The mechanism is running relatively well; given the generous international response, it can avoid widespread or mass famine.

These evaluation and response mechanisms enable fairly large amounts of aid to be mobilised. However, placing unshakeable faith in them is not without risk, and it would be dangerous to rely on them alone:

  • The mechanisms for early warning, assessment of food shortages and distribution were designed for farming regions, and are much less relevant to pastoral areas (where early warning systems do not, in any case, always exist).
  • Emergency food aid is both visible and costly, and eats into other forms of international assistance. As a result, Ethiopia has one of the world’s largest food aid programmes, but the lowest rate of official aid per capita in Sub-Saharan Africa.
  • Short-term emergency operations are not appropriate for chronic or recurring situations.
  • Until 2003, calls for food aid made no distinction between temporary needs (in response to a one-off crisis) and chronic needs (structural food deficits). This means identical responses to fundamentally different problems.
  • The food rations that are distributed are not necessarily suited to real nutritional needs. This gives the illusion of an effective solution, while masking persistent problems.
  • Finally, the process of assessing crops, regional deficits and consequent food aid needs remains cumbersome, complex and plagued by disparate methodologies and regional interpretations. This damages its effectiveness and coherence.

The assessment of the 2003–2004 Meher harvest, accounting for 95% of the total national cereal output, illustrates this incoherence. While official statistics and the joint (FAO/WFP/DPPC/donor) assessment predicted a bumper crop, subsequent analyses of the country’s grain markets by the Central Statistical Authority concur on a harvest on a par with the average for the last ten years. Estimates vary from 8.5m to 11.8m tonnes of cereals, equivalent to a margin of error of nearly 50%. Estimated food aid requirements reflect this uncertainty: the FAO, while speaking of a ‘record crop’, still estimates that 7.2m people might be short of food – a deficit of around a million tonnes. There is, in other words, a fear that an underestimation of needs – perhaps in the hope of restoring confidence in agricultural policy – could lead to catastrophe. In 2003, the area most seriously affected by the nutritional crisis, Sidama zone in the SNNPR, was not one of the regions usually threatened with famine; it had not been considered especially vulnerable in the past, which meant that needs were spectacularly underestimated, and the targeting of beneficiaries was problematic.

Shifting the policy focus

Countrywide food self-sufficiency has long been a primary political and economic objective for Ethiopia: cropping and herding account for around 50% of gross domestic product and 70% of exports; since the mid-1990s, the Ethiopian government has been trying to make agricultural development drive economic development. While Ethiopia has come closer to its overall aim of self-sufficiency, this has not led to an automatic reduction in food insecurity:

  • Annual variations in production are still considerable, reflecting in part the dependence of farming on climatic conditions.
  • These variations cause significant price fluctuations, in the absence of effective market regulation and a transport and storage infrastructure that could allow surpluses to be directed towards deficit areas.
  • The government has encouraged systematic use of inputs. These are supplied on credit, which can cause repayment problems due to the vagaries of production and the absence of a guaranteed minimum selling price. Paradoxically, a good year can undermine producers, causing them to sell at a loss to repay their debts, which will mean reduced production capacity the following year.
  • The lack of security of tenure (the state still owns the land) acts as a brake on rational land management by farmers.

A new approach is needed to the issue of food security, designed to guarantee self-sufficiency at household level, thereby reducing the risk of temporary shortfalls. With this in mind, the Ethiopian government reworked the food security strategy framework in 2002. It has set out four objectives:

Improving food availability by increasing crop and animal production

  • Developing irrigation (3% of cultivated land is currently under irrigation) and water points in pastoral areas.
  • Soil conservation and fodder plantations in pastoral areas.
  • Diversifying crop and animal production, encouraging specialisation in accordance with ecological characteristics (promoting vegetable production in dryland areas).
  • Intensifying cropping through the use of inputs (improved seeds, fertilizer, animal or mechanical traction, pest control).
  • Strengthening extension services.

Improving access to food by increasing farm and off-farm income

  • Developing ‘safety net’ programmes to limit the erosion of productive capacity in the event of crisis.
  • Promoting income diversification.
  • Improving credit and market mechanisms.
  • Improving access to micro-credit mechanisms.

Improving health services

Improving access to land through voluntary relocation out of areas of chronic food insecurity

These new guidelines acknowledge the critical importance of animal production, and the need to develop a coherent policy in this sector. Ethiopia’s livestock population is the largest in Africa, with, according to FAO estimates, more than 40m head of large livestock and the same number again of small ruminants. The food security strategy also makes the link between so-called ‘transitory’ food shortages and ‘chronic’ deficits. It stresses the need to develop a preventive approach in the medium to long term, while reinforcing short-term response capacity.

This development is important, and needs to go further: often, massive so-called ‘emergency’ responses are implemented with no thought for the more fundamental approaches that they end up smothering. Genuine coordination between these approaches, all of which are needed in Ethiopia, is essential.

To take one example: in Afar region, Action Contre la Faim (ACF) is supporting a Community-based Animal Health (CBAH) network. This multi-year project is based on cost recovery, and is designed to operate independently. At the same time, recurrent crises in the region periodically require an emergency response. As things stand, it may be necessary to proceed with a mass campaign of free vaccination and livestock treatment, even though this goes against the cost-recovery and self-financing principles of the CBAH system. However, it is possible to limit the contradictions by establishing strict criteria, common to all agencies in the region, before an emergency situation is declared; restricting the length of the emergency programme to the barest minimum (typically, one month); involving ‘long-term’ stakeholders (the herders from the CBAH system) in designing and implementing the programme; and raising beneficiaries’ awareness of the usefulness of the veterinary service, and the fact that its being provided free of charge is exceptional (not easy, of course). Where these principles are not put into practice, the CBAH system suffers long-term effects: in Afar zone 1 in 2003, for example, massive free drug distributions saturated the market and ruined the CBAH system.

The failure of market mechanisms is also an important factor in food insecurity. At the end of 2000, in anticipation of a bumper crop, the price of grain in the producing regions collapsed to record lows (at times, well below production costs) for over a year. This impoverished producers, who were unable to meet their credit commitments, and made them extremely vulnerable in the next cropping season – which turned out to be a disaster. At the end of 2003, anticipating another bumper harvest and therefore the possible repetition of this process, the government established a floor price slightly above the producers’ average cost price. This appears to be a logical counterpart to a policy of intensified production, since the idea is to ‘guarantee’ small producers a minimum return on their investment. However, the system was not tested this year, because prices naturally held up above the intervention price. According to some international experts, this policy is unrealistic in any case, since the government does not have the means to intervene on an adequate scale in the event of a collapse in prices unless it has support from donors. Donors are, however, unlikely to want to get involved in what looks like a subsidy to Ethiopian farming, even though such practices are common in the West.

In the absence of an adequate transport and marketing infrastructure, a solvent market in deficit areas and government capacity to implement a price stabilisation system, the only alternative would be for Western agencies to buy locally, in surplus areas, in order to distribute in deficit areas. The main donors would have to agree among themselves to do so: the European Union has adopted a local purchasing policy since 1996, whereas the US, constrained by its own legislation, uses its own grain surpluses as aid. Moreover, this would in effect put foreign agencies in the position of acting as regulator of Ethiopian grain markets, perpetuating the dependency of the country’s economy.

The food security strategy also acknowledges the need to provide small farmers with greater security of tenure (without going so far as privatising the land, which remains taboo). This is a major bone of contention between the government and the main international donors, for whom the absence of private land ownership will always be a hindrance.

The final and most controversial aspect of this policy is the voluntary transfer, over three years, of 440,000 households affected by chronic food insecurity. The dramatic failure of similar experiments in the 1980s, as well as comparable projects in other countries, invites scepticism. Political conditions have undoubtedly altered since the 1980s: the government now emphasises the voluntary nature of the plan, the occurrence of spontaneous relocations that need to be supervised and supported and, finally, the intra- (rather than inter-) regional nature of the planned population movements. Above all, the government says that these transfers are unavoidable, inasmuch as the people concerned are now dependent on food aid for their survival, with no real prospect of regaining their self-reliance. Projects have already been launched, with no real support from international donors. While it is far too early to judge this resettlement policy as a whole, as it is designed to last three years and will have an impact over many more, serious problems have already started to appear at certain relocation sites, especially in Oromo region, including health problems, malnutrition and a failure to prepare sites before families arrive. This will probably turn out to be one of the most important humanitarian issues in Ethiopia over the coming months.


The experience of 2003 underlines the persistent need for food aid, and its ability to ward off a major or widespread crisis when it is properly implemented. Yet it also illustrates the weakness of an aid system based on anticipating emergency needs (always a risky and uncertain exercise), and the lack of understanding of the counter-productive medium-term effects of aid. Can the international community confine itself to responding to annual appeals whose main virtue is that they send the mass of Ethiopian farmers back to oblivion until the next warnings of ‘record famine’? It is thought that a ‘hard core’ of around five million people will require food aid in any one year, whatever the climatic conditions. Another two or three million may need food aid, even in good years, because an earlier crisis has affected their means of production. Given population growth of 2.7% a year, recurrent climatic problems and better geographical coverage by early warning systems, the number of beneficiaries can be expected to ‘casually’ rise beyond ten or even 15m in bad years. This is a pattern that urgently needs breaking, not least because, one day, donor fatigue or a less favourable geopolitical context may fatally hinder the emergency provision of food to 15m Ethiopians in dire need.

Jerome Frignetis Ethiopia Desk Officer for Action Contre la Faim. His email address is

References and further reading

Federal Democratic Republic of Ethiopia, Food Security Strategy, March 2002.

SIDA/WFP/EC, Availability of Cereals for Local Purchase, March 2004.

FAO/WFP, Crop and Food Supply Assessment, January 2004.

Ethiopia Network on Food Security report, FEWS NET, February 2004.

Yves Guinand, ‘Livelihood Disruption in Cash Crop and Surplus Producing Areas’, Emergencies Unit for Ethiopia, August 2002.


Comments are available for logged in members only.