Appendix to Chapter 3PEA: The Example of Food Aid in Angola - Oxford Scholarship Jump to ContentJump to Main Navigation
Humanitarian EconomicsHumanitarian Economics$

Gilles Carbonnier

Print publication date: 2016

Print ISBN-13: 9780190491543

Published to Oxford Scholarship Online: September 2016

DOI: 10.1093/acprof:oso/9780190491543.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE ( (c) Copyright Oxford University Press, 2015. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see Subscriber: null; date: 28 September 2016

(p.205) Appendix to Chapter 3

(p.205) Appendix to Chapter 3

PEA: The Example of Food Aid in Angola

Humanitarian Economics

Gilles Carbonnier

Oxford University Press

Food aid has consistently been highlighted as particularly prone to diversion and political instrumentalization. As an illustration of the applicability and relevance of political economy analysis (PEA) in the humanitarian sector, I provide in this appendix a concrete example of PEA being carried out in the context of a large-scale emergency food assistance and agricultural rehabilitation programme in Angola.

When hostilities resumed between the Government of President Dos Santos and the Uniao Nacional para a Independência Total de Angola (UNITA) in December 1998, the country witnessed a surge of internally displaced people (IDPs) fleeing rural areas towards government-held enclaves such as Huambo and Kuito on the Planoalto. The 1999 spring harvest had been very poor, not least because farmers opted for premature harvesting to avoid the risk of later plunder by warring parties. In August 1999, the incidence of acute malnutrition surged to an alarming rate of 40 per cent, which was comparable to what was registered in Somalia during the deadly 1992 famine. The International Committee of the Red Cross (ICRC) and the World Food Programme (WFP) coordinated their airlifted food aid operations, the former servicing some 330,000 beneficiaries in Huambo and its outskirts. In less than a year, acute malnutrition dropped to 3 per cent.

At the end of 1999 I was asked to assist the ICRC nutritionist and agronomist in assessing the adequacy of the relief operation from a socioeconomic viewpoint. We carried out a political economy analysis: for each phase of the food aid operation we examined the interests, objectives and potential strategy of the major stakeholders involved and how the ICRC should address them. This analysis is summarized in Table 1, where each row refers to one stage in

Table 1: Relief Access Mapping—Food Aid in Angola (1999–2000)

Stages of the relief operation

Main stakeholders involved

Stakeholders’ interests and likely strategies

Risks from a humanitarian point of view

Preventive and corrective measures by relief agencies

Initial need assessment

Central and local government officials

Avoid a food crisis and ensuing turmoil, yet without having to mobilize domestic resources

Aid fungibility

Independent rigorous need assessment, cross–checking with local traders

Get political reward for getting food distributed

Discharging the authorities from their obligation vis–à–vis the civilian population

Persuade the authorities to protect and assist the civilians


Get continued food assistance

Aid dependence, negative spin–off for local farmers

One–off, simultaneous distribution of food with seeds, tools and fertilizers


Attack convoys to stop and divert food aid

Security of staff and facilities

Airlifted operation, planes spiralling up & down on

landing strips in enclaves Parallel food distribution in rebel–held area?

Funding (cash & kind)


Get rid of food surpluses; promote the marketing of GM (genetically modified) maize seeds

GMO (genetically modified organisms) unwillingly disseminated; dependence on imported seed varieties

Milling the imported maize for consumption; buying local maize seed varieties for planting

Beneficiary selection and food distribution modalities


Select beneficiaries among “sympathisers”. Get political reward for the distribution

Fraught targeting, not based on actual needs

Independent selection of beneficiary list. Direct distribution to individual beneficiaries

Security forces

Looting part of the food distributed for own consumption and monetization

Security risks and nutritional status of the beneficiaries

Escorting beneficiaries back home; advocating for adequate wages to be paid to security forces

Procurement, logistics, recruitment, sub–contracting

Local labour

Competition to get access to jobs

Unbalanced recruitment antagonizing groups against each other or against relief organizations

Context–sensitive recruitment procedures, under expatriate responsibility

Traders and locals businesses

Gaining new market opportunities, not losing existing ones

Security risks

Dialogue with business people; corroborate if need assessment is validated by market signals

Monopolists (often a coalition of political and economic elites)

Preserving monopoly rent (e.g. microcredit scheme for buying seed and fertilizers)

Security risks for humanitarian workers

Careful monitoring and communication with key actors (e.g. this is a one–off distribution)

Project monitoring and evaluation


Keep getting free seeds and fertilizers while keeping food prices high

Discouraging farmers from planting as a result of depressed food prices

Close monitoring of evolving nutritional situation and access to food. Agricultural rehabilitation programme

Logistics providers

Continued business


Continued business and/or stepping in to import the food once free distribution ends

Coalition of actors pushing for protracted assistance and renewed food aid

Monitoring and evaluation of the humanitarian, economic and political economy effects of the relief operation

(p.206) (p.207) (p.208)

(p.209) the relief operation and the columns address the interests and strategies of the key actors as well as the preventive and corrective measures that relief agencies could take. Such a Relief Access Mapping (RAM) table permits humanitarian workers to collectively explore the political economy of an aid project both within field offices and in interactions with headquarters. It further offers a relatively simple yet powerful tool for highlighting potential security risks.1

Table 1 presents the RAM that summarizes the main issues raised as part of our PEA related to the 1999–2000 food aid operation in Angola

The issues raised in Table 1 are self-explanatory, with four issues that deserve further elaboration:

  • Funding: since the food to be distributed was primarily maize shipped from the United States, we discussed the risk of recipients keeping some of the maize beans for planting instead of eating, and the ensuing risk of disseminating genetically-modified (GM) maize on the Angolan Planoalto. The decision was taken to mill all the imported maize in order to distribute only fuba or maize flower for consumption, even if this slightly delayed the distribution and made the operation costlier. At the same time, the ICRC purchased local maize seed varieties to be distributed with agricultural tools and fertilizers for the next planting season.

  • Distribution: some food recipients reported that they were the victims of extortion at checkpoints when heading home from distribution sites. The first reaction was to have ICRC expatriates escort them through the checkpoints, which ensured safe passage. Yet, some food recipients later reported that local security forces intruded at night to levy dues directly from their homes or shelters, with serious security risks for the recipients’ families. It became obvious that the security forces would continue to loot as long as they did not get properly paid by the state to sustain themselves and their families. This prompted the ICRC to intervene directly in the capital city, Luanda, with the relevant ministries. The government released its decision a few months later to significantly increase the remuneration of the security forces.

  • Procurement/Logistics: from the initial need assessment phase to the final ex-post evaluation, regular exchanges with local business people provided particularly useful insights. The latter started by expressing outright support for the free distribution of imported food into the government-held enclaves on the basis that the food recipients were not among their potential clients. In other words, they were not solvent. Distributing food for free would thus not compete with local business interests. This reinforced (p.210) our conviction that food distribution was needed. Local business people insisted that we keep paying our staff in US dollars. As aid agencies were the largest employer in town, this was critical to sustain the demand for the consumer goods that local traders imported into the enclaves. The only reservation came from the local politico-economic elite involved in (micro)credit institutions. Some of them were used to extending loans to farmers allowing them to purchase agricultural inputs before the planting season. They were also involved in selling seeds. Explaining that the free distribution of seeds, tools and fertilizers was a one-off donation to restart agricultural activities helped to ease the tensions and avert potential attempts to derail the seed distribution programme.

  • Monitoring and evaluation: we distinguished between three categories of outcomes when putting the monitoring and evaluation system in place:2 (i) the humanitarian effects of distributing food rations related to the evolving rate of acute and moderate malnutrition; (ii) the economic effects related to the local price fluctuations of maize and fuba together with that of other staple foods. We also thought it useful to monitor any potential surge in transport and warehousing prices as well as changes in wages in specific labour market segments such as daily labourers; and (iii) the political-economy effects as detailed in Table 1.


(1.) The Relief Access Mapping used here is adapted from the RAM framework suggested by Philippe Le Billon in: ‘The Political Economy of War: What Relief Workers Need to Know’, Humanitarian Practice Network Paper 33, London: ODI, July 2000, p. 20. On the political economy of food aid, see also: David Keen, The Benefits of Famine: A Political Economy of Famine and Relief in Southwestern Sudan, 1983–1989, Princeton: Princeton University Press, 1994.

(2.) We followed an approach that has actually been advanced by Frances Stewart and Emma Samman, ‘Food Aid during Civil War: Conflicting Conclusions Derived from Alternative Approaches’ in F. Stewart and V. FitzGerald (eds), War and Underdevelopment, Vol. 1, Oxford: Oxford University Press, 2000, pp. 168–203. The authors tested the approach in a comparative case study looking at the three types of impact of food aid in Afghanistan, Mozambique and Sudan to get an overall sense of the positive or negative balance. (p.258)