Food aid logistics and the southern Africa emergency

The recent protracted crisis in southern Africa stretched the ingenuity and capacity of the international community.

In the six countries of the Southern African Development Community (SADC) - Mozambique, Malawi, Zambia, Zimbabwe, Lesotho and Swaziland - a jigsaw of accumulative factors were to bring about a heightened crisis: the volatile mix of drought, floods, disruptions to commercial farming, the absence of effective food security and governance policies, depletion of strategic grain reserves, poor economic performance, foreign exchange shortages and delays in the timely importation of maize. The sub-region has the worst HIV/AIDS prevalence rates in the world - a major contributing factor towards household food insecurity that will have long-term development implications.

One might have expected a fairly straightforward response to the crisis given that (a) the 1992 food crisis had provided useful lessons and helped build early warning capacities in the region; (b) the regional infrastructure was relatively robust - good roads, rail and port facilities; and (c) in spite of political obstacles, significant amounts of surplus maize were still being grown and governments gave early indications that they would meet domestic shortfalls through commercial imports and subsidies. Full advantage was taken of the strength of the commercial sector in southern Africa. The World Food Programme (WFP), by far the largest food aid delivery agent, was able to outsource the handling at the ports and management of rail transport in South Africa, transhipment points and warehouses.

What was not foreseen, however, was the series of extraordinary events surrounding the GMO (Genetically Modified Organisms) issue. It is this issue in particular that is examined here, for it was to have both positive and negative consequences for the relief operation, with a steep learning curve for those agencies adapting their approaches to unfolding events.

Attitudes towards GM

In mid-July 2002 UN agencies launched a special appeal for Southern Africa - OCHA's 'Regional Humanitarian Assistance Strategy' - requesting some US$611 million of which 90% comprised food aid. WFP led inter-agency coordination, information exchange and advocacy as WFP's Executive Director served as the UN Secretary General's Special Envoy for Humanitarian Needs in Southern Africa. The WFP appeal alone initially involved the provision of 992,459 tons of food over nine months, with some additional tens of thousands of tons to come through NGOs.

As with most large-scale food interventions, the US Department of Agriculture (USDA) was to provide the majority of in-kind contributions in the form of whole maize. What they had not anticipated was the rejection of this food aid by some governments because it was genetically modified. It was difficult to distinguish political manipulation and obfuscation from genuine environmental, health and economic concerns. There were political interests on both sides of the debate. The USDA clearly did not want to create a precedent for governments to reject its food surplus exports as aid. WFP's official policy is essentially one of neutrality, stating that the acceptance or rejection of any such food donations is the prerogative of the recipient government. A UN joint statement of 27 August 2002 on the use of GM foods in southern Africa went further by indicating that no scientific evidence is yet available to suggest a risk to human health from GM foods. On the issue of potential spillage and cross-fertilisation, the statement leaves this to the judgment of recipient countries.

Though initially in favour of accepting GMOs, Zambia began to take a more hard-line attitude in 2002 just as the international community was gearing up for a major food aid intervention. By mid-August it had banned all imports of GM products, including those on the high seas already committed to the aid operation. WFP and other major donors were required to withdraw all existing stocks from the country at considerable cost. This included food destined for the 130,000 Angolan refugees in camps (though the government would accept milled maize for these people). Meanwhile, Kenya and Tanzania offered 'natural' maize to offset any further deficit once Zambia had purchased 300,000 tons from South Africa.

The Zambian decision had something of a domino effect. Bureaucratic delays and procrastination meant that quite significant amounts of GM maize had already been distributed in all countries but by December 2002 the following positions were confirmed:

  • Zimbabwe banned all GM maize grain (unmilled) but was willing to accept some quantities for milling in Bulawayo prior to distribution (milled maize has neither re-planting/cross-fertilisation risks, nor the risk of consumption by cattle(1), but obviously still does not address potential human health risks).
  • Mozambique banned grain but accepted in-country milling, provided extra funds were made available to meet this cost.
  • Lesotho and Malawi in theory did not accept grain but existing and in-transit supplies were accepted.
  • Swaziland was the only SADC country to accept GM maize.


Implications for logistical operations

i. Shipments and milling requirements

The immediate consequences of the crisis were a disruption of aid supplies to vulnerable populations for at least a month. Large shipments of GM maize were stranded at the ports of entry; if these were in areas of high humidity, milling problems and wastage occurred later. WFP had to make immediate arrangements to mill large quantities of GM maize in South Africa. Previously mothballed mills were re-opened but owners insisted on some kind of guarantee of forthcoming quantities, always difficult in an international appeal with a staggered response. Milling extraction in South Africa (where the majority was done) involves 25% reduction in the total cereal available for distribution from this source, since the offtake and some of the maize meal are taken as payment. Finally, the additional handling, superintendence, tallying and transport costs required to move GM maize and resultant maize meal in and out of mills further increased already high overland costs incurred by aid organisations.

ii. Local purchases

Two positive results came from the need to mill GM maize. It gave WFP an unexpected opportunity to fortify the maize meal at the mills, so meeting micro-nutrient needs of many beneficiaries in the region. Perhaps most importantly, it triggered the largest local and regional procurement of food in WFP's history, using cash that normally would not have been acquired for an operation of this size. Local purchases have the immediate advantage of timeliness and a stimulus to local markets - no long wait for port dispatches, and an increased use of local suppliers and transporters.

Much of the food was procured in South Africa. Compared to the previous 12-18 months, maize prices almost doubled in South Africa in November-December 2002, presumably influenced by expectations of high demand in neighbouring countries due to crop shortfalls. The depreciation of the rand against the dollar also became a key factor behind price rises since maize is bought in dollars. Although for its part, WFP usually bought in small lots as a deliberate policy to avoid adversely affecting the market, maize prices nevertheless rose to $195/ton. This, and pipeline delays on international deliveries, meant that planned distributions were not always achieved. Ideally, WFP would like to purchase less expensive maize from countries such as China rather than regionally produced white maize sold at premium rates but limited and late availability of funds meant this was not possible in the given timescale.

iii. Widening of donor base

WFP paid for over 45% (332,000 tons) of its food commodities in the southern Africa region (as opposed to in-kind contributions, primarily from the US, that in many emergencies elsewhere in the world would have accounted for as much as 70% of commodities). An unusual array of non-traditional donors was found - more than 40, including from developing countries.

In a recent book Edward Clay has pointed to the gradual erosion of the multilateral character of WFP(2) as one donor in particular (the US) dominates the global food aid arena. Could the southern Africa emergency have been a turning point? Perhaps, though not without a cost: with cash donations, many new donors do not provide the same generous overheads per ton of purchased food as the US in-kind food allows. Also, economies of scale - precisely the strength of WFP - may be compromised by having a large number of small and dispersed markets to deal with. There can be several months between the confirmation of a pledge, the release of money, and the purchase and delivery of food. In this operation distribution targets were below 50% in most countries for the first five months.

Lessons and questions

The response to the GMO crisis was greatly facilitated by the early setting up of an impressive regional management and logistics centre in the WFP regional office in Johannesburg. It included a web-based information system (ReLogS - 'Regional Logistics - Southern Africa') that was updated daily with pipeline information, country specific situation reports, port operation and other relevant logistics and programme information. Much appreciated by all stakeholders, this brought a welcome degree of cohesion into a complicated operation.

One clear lesson was the need to develop a more reliable and comprehensive way of assessing the 'food gap'. Currently, the UN Food and Agriculture Organisation (FAO) and WFP do joint Crop and Food Supply Assessment Missions (CFSAM). CFSAMs were undertaken in all countries in May/June 2002. They are based on four sets of data: (a) projected national production of key crops; (b) what percentage of these will be consumed or exported; (c) commercial imports; and (d) levels of government food aid from existing stock (for example, grain held in reserve by the Grain Marketing Boards). The resulting deficit becomes a guideline for quantities of food aid required from external sources or purchased regionally. All national figures on production, consumption, importation and government aid rely upon accurate forecasts by government ministries. This has two basic weaknesses: first, the capacity of some Ministries of Agriculture has declined in recent years; and second, there can be important differences between national macro figures and provincial/local differences that often are not reflected in the balance sheets of the CFSAM.

Apart from deficiencies in government data, there were also some shortcomings in the way internationally supported early warning systems produced forecasts in 2002. In Malawi, for instance, there was a wrong assumption that the household consumption of locally produced roots and tubers would compensate for cereal deficits. This did not happen on the scale predicted, so the overall balance sheet forecasts were underestimated. Interestingly, it can also work the other way: in Lesotho in July 2003, there were reports of beneficiaries not turning up to food aid distributions since in some areas their harvest had been better than predicted.

The dominance of the food aid operation and the projected needs foreseen by the unique Vulnerability Assessment Committees (VACs)(3) set up in each country may have overshadowed more fundamental questions. In Zimbabwe logisticians faced problems of inflation, fuel shortages and restrictions on hard currency that worked against the smooth running of an operation of this scale. Some governments, notably Zimbabwe, were only too happy to accept the mantle of 'drought emergency' to divert attention from serious governance, economic and policy failures. In Malawi donors were subdued by the extraordinary sale of most of the Strategic Grain Reserve in late 2001 - a combination of ill-conceived IMF advice and domestic corruption. And still today little is known of the extent of informal cross-border trade that would have cast a different light on the grain import/export balance sheets that determined exactly what the food 'gap' was.


Southern Africa faces a protracted chronic and structural - i.e. developmental - challenge. Declining development indices combined with the HIV/AIDS 'permanent emergency' mean that short-term food aid has a limited impact. The GMO crisis and the unusually large regional purchases it prompted have highlighted the need for:

  • the larger international agencies to develop new set of analytical skills to better understand both the economic variables that determine grain supply and demand, particularly in the informal market, and the decisions that farmers make over whether to sell or retain their produce
  • understanding the effects of illegal cross border trade in state-dominated markets such as Zimbabwe
  • greater inter-agency knowledge sharing on the potential adverse effect that aid demands have on grain prices and the commercial transport sector
  • close liaison between high level advocates such as the Special Envoy, whose role it is to negotiate government concessions over such things as import duties and border controls, and those logisticians who face day-to-day difficulties translating these policies into practice with sometimes unhelpful local authorities.


The lessons from the emergency, the greater emphasis given to livelihoods approaches and the attendant programme adaptations already underway should ensure more sustained attention in future to problems facing the region.


Jon Bennett is Director of Oxford Development Consultants and has recently led a 'real-time' evaluation of the emergency response in southern Africa. Email:


  1. This was only a perceived risk, since the EU - a major importer of dairy and meat products from southern Africa - declared that cattle that consumed US maize would still be acceptable.
  2. Edward Clay, 'Responding to Change: WFP and the Global Food Aid System' in Food Policy Old and New , Edited by Simon Maxwell and Rachel Slater, Development Policy Review 21 (5), Overseas Development Institute, London (forthcoming).
  3. See text box below.
  4. See 'Structural damage: the causes and consequences of Malawi's food crisis', World Development Movement, October 2002,


Vulnerability assessment

SADC member countries established the Vulnerability Assessment Committee (VAC) in 1999 with a regional office in Harare. During the 2002/3 regional emergency, its resources and staff were boosted by the secondment of UN staff, involvement of NGO partners and additional financial support from USAID, DFID and the WFP Regional Emergency Operation.

The VAC keeps abreast of and encourages coordinated development in the field of vulnerability and livelihoods assessment in the SADC region. It collates and coordinates data from existing sources and supplements these with additional information derived from on-site periodical surveys. In each country, a national VAC comprises government and partner agencies, with field teams drawn from relevant government departments, NGOs and some UN staff. Advice on methodology and training is provided by the regional VAC team based in Harare, which is also responsible for data collation, analysis and publication.

Throughout the emergency, 'rolling assessments' have been undertaken to regularly gauge needs on the ground in order to prioritise districts according to food aid needs (and rations) and to derive a national level food assistance total (indicating how long the requirement will stand before the situation may change again). VACs are supplementary to the FAO/WFP Crop and Food Assessment Missions (CFSAMs) in that they provide sub-national food aid targeting requirements, including a breakdown of socio-economic groups and special needs targeting (e.g. HIV/AIDS, child supplementary).

Translating the VAC updates on targeting priorities and overall food aid requirements into programme action has presented a challenge. Limitations in NGOs' capacities as well as differing priorities of village/district committees and redistributions at village level make quick adjustments to a revised needs analysis difficult.

There was initially a hope that the VAC methodology might be adapted to move from 'needs assessment' towards an impact analysis of the emergency response. In the event, although it was possible in 2003 to include a few questions that would help understand the effects of the emergency response, the VAC was not actually an appropriate vehicle for impact assessment. By April, there was already the broadening of the assessments to include HIV/AIDS food security linkages, but further expansion may have reduced the quality of the data and overstretched the capacity of the VACs.

Maintaining the high level of inter-agency consensus on VAC findings will depend on continuing investment and government commitment. There is a risk that too many expectations will be made of the VAC.

For more information, see:



Opinions in FMR do not necessarily reflect the views of the Editors, the Refugee Studies Centre or the University of Oxford.
FMR is an Open Access publication. Users are free to read, download, copy, distribute, print or link to the full texts of articles published in FMR and on the FMR website, as long as the use is for non-commercial purposes and the author and FMR are attributed. Unless otherwise indicated, all articles published in FMR in print and online, and FMR itself, are licensed under a Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. Details at