Emergency food-based programming in urban settings

Summary of published research¹

Children attending Stara School, Nairobi, that receives WFP food support.

The Food and Nutrition Technical Assistance 11 (FANTA-2) Project has recently published a paper to provide technical information and lessons learned to support USAID and its partners to design effectively and implement emergency food assistance programmes in urban and peri-urban settings. The USAID Office of Food for Peace (FFP) food assistance programmes have traditionally focused largely on rural areas, offering a large body of literature on best practice. Similar information on urban food assistance programming has been limited, particularly in emergencies.

The paper sets out a number of factors that may differentiate urban and rural crises resulting in a different set of response criteria and mechanisms. These include:

• Urban-rural differences in food security.
• Food consumption and access - food is generally the largest expense category in the budget of the urban poor and the main source of food insecurity amongst urban populations is food access.
• Labour market - the urban poor mostly earn cash from low-paying unstable jobs in the informal and formal sectors, with populations displaying a higher degree of economic inequality than rural areas.
• Social networks and gender - urban social networks are based on political, religious and economic, as well as ethnic, affiliations. Although ethnicity is the dominant factor in determining residence and settlement in some cities, as a general rule, groups tied together by these factors are less geographically circumscribed in urban than in rural settings.
• Public health and care practices - urban populations as a whole make greater use of health services than rural populations.
• Institutional and stakeholder environment - there is a broader range of stakeholders that can serve as partners in urban areas than in rural areas.
• Socioeconomic status - a given population's socioeconomic status is usually higher but more variable in urban areas, with remittances a critical source of income.
• Property issues - urban areas in low-income countries often include unplanned settlements or slums, whose residents are frequently at high risk of food insecurity.
• Rural-to-urban migration and connections - urban and especially peri-urban residents often retain active economic and social links with rural areas. Health care in cities may also be a draw for rural to urban migration, particularly for HIV services.

Planning food assistance in urban emergencies

The paper sets out how to plan food assistance in urban emergencies and specific issues that relate to the urban environment.

Assessment

There is no internationally accepted standard guide or instrument for urban emergency needs assessment. The authors suggest that time and resources permitting, a rapid nutrition survey can be conducted to help provide a representative picture of groups of concern. However, urban under-nutrition prevalence estimates often mask significant variation among socioeconomic groups. Assessing food security in urban contexts is also a challenge. Assessments should aim to capture impact of the crisis on expenditure on basic needs and household income because urban populations tend to access most of their food from the market. There are key contextual issues to consider. Population density and poor sanitation elevate the risk of communicable disease epidemics, including acute diarrhoeal disease. HIV is also more prevalent in urban populations, increasing nutritional requirements and morbidity risks during a crisis. Exclusive breastfeeding rates are lower in urban than rural settings, so the risk of contamination of food and water is elevated in urban emergencies.

Targeting

Deciding what targeting approach to use usually depends on the proportion of the population that needs assistance, the type of programme contemplated, trade-offs between targeting cost and targeting accuracy, and the feasibility of targeting options. Considerations include:

• Proportion of population to receive assistance - if the large majority need assistance and resources support a large scale response, then it is not cost-effective to select households and individuals.
• Type of programme - some programmes require targeting eligible institutions and providing assistance to all individuals at those institutions. Others, like targeted distribution of 'take home' rations, require identifying households that meet certain social or economic criteria.
• Targeting costs and accuracy - targeting generally entails a trade-off between costs and accuracy.
• Feasibility of targeting approach - community based targeting requires accountable and strong community structures, transparency and high community engagement.

Exit strategies

Although most urban emergency food assistance programmes will aim to address transitory food insecurity within a short (e.g. one year) time frame, exit strategies must still be identified from the outset. Many of the factors contributing to urban food insecurity - including rising food prices and climate changerelated shocks - are expected to be medium to long term. This underscores the need to address underlying causes, as well as to establish ways to support these populations through national structures (e.g. social safety nets) wherever possible. An urban emergency food assistance programme's exit strategy should include components such as graduation and exit criteria, an exit timeline, benchmarks and activities required to meet those benchmarks, a monitoring and evaluation strategy, and a sustainability strategy.

Security

Population density in urban areas heightens the need for security and crowd management at sites where food is distributed.

Institutional partners and stakeholders

Urban settings offer several opportunities to partner with local institutions and stakeholders. Two types of stakeholder are more prominent in urban settings: central level public service delivery institutions (e.g. health, education, nutrition extension) and private sector actors. This is important to consider when developing an exit strategy and the handover of programme services. Urban authorities are more able to monitor and influence programme activities in urban settings, underscoring the need for an active partnership, transparency and even legal agreements to document plans, roles and responsibilities related to the programme. Unless urban food assistance programmes work solely through national systems, e.g. school feeding, MCHN (Mother, Child Health and Nutrition) clinics, they frequently require negotiating with the urban private sector. For example, access to land is required for wet or dry ration distribution sites, while access to infrastructure is required for food for work.

Urban food assistance programme options

The remainder of the FANTA-2 report describes and analyses eleven types of urban food assistance programme options. These are:

• Targeted household food distribution
• Food for work
• Food for training
• Wet feeding programmes
• Community-based management of acute malnutrition programmes
• Supplementary feeding in MCHN programmes
• Institutional feeding for street children, orphans and other vulnerable children and other vulnerable groups
• School feeding programmes
• Food support to child care facilities
• Market assistance programmes
• Support to national strategic food reserves

For each of these interventions, advantages, disadvantages and programming issues in an urban context are considered. Programming issues include targeting, long-term developmental considerations, specific programme requirements, risks and potential pitfalls, monitoring and evaluation considerations and exit strategies. Urban country examples of each type of intervention are also given.

For the purposes of this summary, we only highlight key points raised in the paper relevant to implementing these interventions in an urban context. We also select one case example for each type of intervention.

Targeted household food distribution

This involves the direct distribution of dry takehome food commodities, usually to pre-identified target groups. It typically targets a larger population than just the traditional category of vulnerable individuals, e.g. elderly or female headed households. Objectives include meeting immediate food needs of households unable to access their normal sources of food, preventing famine, protecting livelihoods and/or supporting livelihood recovery and freeing up income for other household needs. Urban targeted household food distribution poses a number of challenges, including defining target groups and identifying eligible beneficiaries (the risk of inclusion error can be high especially with urban political and social elites), exclusion of non-beneficiaries in densely populated areas, defining household in a way that encompasses the diversity of household compositions seen in urban settings and assisting a large and relatively mobile population.

Ration planning should consider what staple foods are most appropriate for urban populations. These populations often have diverse diets, less time to prepare distributed commodities, and may show higher rates of specific chronic illness, e.g. HIV. Urban households are also heavily cash dependent and some degree of monetisation of take-home rations should be expected to pay for other household necessities like rent and fuel. The exit strategy of an urban targeted household food distribution is particularly challenging, given the inequality and structural poverty in urban centres. Where possible, phase-out will be contingent upon the re-establishment or expansion of income and food sources among beneficiaries and affordability of local staples, rather than production and the agricultural calendar as in rural settings. Finally, it is important to be aware that distribution sites can help link beneficiaries up with other social or economic services to which marginalised groups may have had poor access. In 2008, the World Food Programme (WFP) distributed targeted household rations (wheat) to urban and semi-urban households most affected by rising wheat flour prices in Afghanistan. The programme targeted 1,147,000 chronically poor people, especially female-headed households, disabled heads of households with unreliable income, poor households with more than 9 family members and only one income earner, and some recently deported poor households from Iran. Beneficiaries were selected by non-governmental organisations (NGOs), government institutions and community councils.

Food for Work (FFW)

Urban FFW programmes are public works programmes that pay workers with food rather than cash. Given the highly monetised urban economy, such schemes naturally tend to favour the most food insecure who will self-select into them. Projects normally focus on urban infrastructure - most notably public water, sanitation and hygiene services - thus providing important public benefits.

There are numerous challenges to urban FFW. Activities usually involve issues of private property and civil planning before and after the project, requiring complex consultations and arrangements with a range of stakeholders. Speculation can follow infrastructure improvements, leading to the eviction of the very people the project aims to assist. Special attention is required to protect beneficiaries from possible effects of raised rents and property values following infrastructure improvement. Working through municipal government structures is essential to clearly delimit areas in which the project will take place, where benefits will be expected and what community support is required. Targeting residents of informal settlements and slums can be sensitive politically, especially where construction in those areas is in violation of municipal regulations.

CARE implemented an urban food for work programme in 25 urban slum communities (kabeles) of Addis Ababa from 1997-2001. The project focused on road and sanitation infrastructure and the majority of participants were women.

Food for Training (FFT)

In FFT, beneficiaries are given a food ration in exchange for their participation in an activity designed to impart knowledge or skills that can generate income. Food can be provided as take home or on-site feeding or a combination of the two. Food is used to compensate for the time spent in training and the opportunity cost of other economic activities forgone. FFT can support recovery of crisis-affected communibeneficities, such as for ex-combatants newly integrated into urban and peri-urban communities. It can also promote community self-management, although government systems should not be duplicated. FFT projects in urban areas can build capacity of small-scale traders and informal entrepreneurs. However, effects of FFT on rural-to-urban migration should be assessed.

World Vision's Food Aid Supporting Transformation in Urban Population programme ended in 2008. It focused heavily on FFT with an emphasis on health, nutrition, water and sanitation in East Jakarta and Surbaya. The programme had about 4,300 FFT beneficiaries in 2007.

Wet feeding programmes

These programmes distribute prepared, cooked foods to beneficiaries for on-site consumption. They may be appropriate under very specific circumstances, such as where a distribution of take-home rations is not feasible due to physical insecurity. It will be necessary to situate wet feeding sites throughout the urban area to minimise the risks of over-crowding or violence and other antisocial behaviours, as well as to reduce the distance that programme beneficiaries have to travel for meals. As urban wet feeding is management intensive, it requires working through local partners. This can mean a longer start-up time if their capacity is not already strong. In densely populated urban areas, wet feeding can encourage overcrowding around the distribution site, requiring crowd control and management of hygiene and sanitation risks. Wet feeding can induce rural-tourban migration if services are not in place to address acute food insecurity in surrounding rural areas. Working with highly mobile urban populations can be a challenge for monitoring and evaluation.

SAACID, a Somali NGO, worked with the Danish Refugee Council to run a WFPsupported blanket wet feeding programme in greater Mogadishu in 2008. The programme aimed to help the most marginalised and vulnerable internally displaced people within the city, as well as the urban poor who lost access to work due to conflict. The project expanded from 10 to 16 sites and from 50,000 to 80,000 meals per day in Mogadishu. Beneficiaries were self-selected.

Community-Based Management of Acute Malnutrition (CMAM)

These programmes address acute malnutrition through inpatient care (for children with severe acute malnutrition (SAM) with complications and infants below six months of age with visible SAM), outpatient care (for children with uncomplicated SAM), and community outreach.

For urban families with economically active adults, CMAM reduces the time that caregivers must spend in inpatient facilities during stabilisation and recovery. Referral for health and other services from a CMAM programme, when necessary, is generally easier in urban than rural settings due to proximity of health centres. However, CMAM requires a strong community outreach system to support timely referral and coverage that can be challenging in urban environments - especially in unplanned settlements. In order to minimise the risk of overburdening the existing system, additional nutritional centres may need to be constructed. In urban areas, referral centres may be urban tertiary hospitals, which are frequently overwhelmed with cases of SAM. Because urban populations often seek health and nutrition related services from a large variety of sources, e.g. traditional healers, private and public sector, it is especially important to establish collaboration with these providers and ensure that core messages related to nutrition are consistent among these actors. Since income insecurity in urban areas is more related to economic volatility than agro-climatic seasonality, predicting and planning for increases in case load can be more challenging. Another potential issue is that because paid outreach workers and community volunteers are more likely to come from the same community in urban settings and may be social peers, it might be more problematic to have both categories of workers within the same programme (one of which is paid and the other unpaid).

Valid International (with Concern support) implemented the scale-up of CMAM in Lusaka, Zambia in partnership with the Ministry of Health.

Supplementary Feeding in Maternal and Child Health and Nutrition Programmes (MCHN)

Food can be distributed through MCHN programmes to meet a number of objectives: prevention of malnutrition in at-risk groups, recuperation from malnutrition of beneficiaries determined to be malnourished, and as an incentive to participate in programme services. Targeting of beneficiaries for food support through MCHN programmes in an urban emergency would include preventive models targeting pregnant and lactating women and children 0-24 months of age. A risk factor may be that if rural areas are underserved, urban programmes could become a magnet.

A 2006-9 WFP programme in Angola includes supplementary feeding for about 15,000 pregnant and lactating women through health centres in food-insecure areas.

Institutional Feeding for Street Children, Orphans and Vulnerable Children (OVC) and other Vulnerable Groups

This refers to a broad range of interventions all characterised by provision of food resources to urban institutions that provide services to street children, OVCs and other vulnerable groups. Facilities may include orphanages, drop-in centres for street children where they receive basic services, hospitals, long-term care facilities/ hospices. Food can be given to support food security and vocational and life-skill training.

In emergency situations, these sites can be used to expand rapidly food support to vulnerable groups and also offer the opportunity to link beneficiaries with other services. However, programmes can become quickly overwhelmed by rapid increases in demand and like all wet feeding, appropriate sources of cooking fuel and safe water must be found. Long-term development impacts are few unless the support is linked to other capacity strengthening initiatives. A potential undesirable consequence of programmes is to act as incentive for poor or food insecure families to place children who are not orphans in orphanages. It also risks undermining traditional community based mechanisms for supporting OVCs. Also, in lowincome countries, these facilities frequently receive no government funding so alternative funding is required. This is a key issue to consider at the outset of the programme to avoid building institutional dependency on food assistance.

CRS has provided food to orphanages and other OVC facilities in Haiti. CRS provides the dry commodities which are then prepared for wet feeding by facility staff.

School feeding programmes

These programmes distribute food to schools for on-site wet feeding. They aim principally to support food access and prevent deterioration of nutritional status in children of primary school age, and also to boost school attendance and attainment.

School selection in urban areas is challenging because the schools are frequently in close proximity. Selection must be justified in a clear and transparent manner to minimise conflict with neighbouring schools that are not covered.

From 2004-8, WFP implemented a school feeding programme in the peri-urban slums around Nairobi to provide a dietary support for all children attending school to combat short term hunger, improve attention span and cognitive ability and provide an incentive to girls and orphans to enrol in, attend and remain at school. The daily food basket consists of 150g of maize, 40g of pulses and 5g of vegetable oil per child.

Food support to child care centres

These programmes aim to alleviate poverty by providing working parents with low-cost child care to facilitate the development of their young children and to promote community participation in the overall development of children. As with school feeding programmes in urban emergencies, child care feeding may be implemented where there is an existing and well established institution already in place and where additional support in the form of food provision for the crisis-affected community has been determined to be feasible and appropriate. Lack of child care options may represent a major obstacle to achieving household livelihood and food security among urban women, especially women heads of households. The cost per child is relatively low, ranging from US$ 0.58 per child per day in Columbia to US$2.15 in Bolivia. Disadvantages include the fact that each child care centre can only serve a limited number of children to be effective, so that many centres are needed to reach scale increasing logistical complexity. Another difficulty may be that women who often spend very long hours away from home between commuting and work may have difficulty complying with programme participation requirements.

In Guatemala, the Hogares Communiatarios began with donated food (including Title 11 food aid through Catholic Relief Services). Positive results were obtained on household income, child nutrition and psychosocial development. The value of the programme and its popularity among the urban poor has motivated subsequent governments to maintain the programme since its inception in 1998.

Market Assistance Programmes (MAP)

This programme provides food to food insecure households through targeted market sales at subsidised prices. MAPs distribute food through retail outlets and existing traders and use less preferred commodities to assist in self-targeting. Subsidised staple commodities can also be sold through a set of programme vendors to beneficiary households identified to be eligible, i.e. where the programme is not self-targeted.

There are many advantages of these programmes. They can stabilise volatile food prices by infusing cheaper foods into the urban markets. If food insecure households tend to use identifiable vendors, these types of vendors can be specifically enrolled in the programme. Because MAPs work through existing markets, they are easier to undertake than direct distribution which usually requires setting up a separate distribution system. Furthermore, MAPs are usually self-targeting and may be more costeffective than FFW or targeted household distribution because of the reduced cost of distribution and the possibility of earning revenue that can help cover costs.

MAPs do pose a number of potential risks and pitfalls. Subsidies may stimulate rent seeking behaviours among politically and economically advantaged groups, such as traders and government employees. Existing market stocks, purchased earlier at higher prices by retailers, will dampen the initial effect of the subsidy as traders will want to reduce their loss by passing it progressively to consumers.

Subsidies require market monitoring to detect market distortion, while the implementing agency must have the capacity to monitor the target population to determine if the target beneficiaries are being reached. There are also exit strategy issues. The main obstacle to phasing out food subsidies are political resistance as people begin to feel entitled, particularly where there is concern that the poor will not be able to purchase the staple food at free market prices.

Since 2003, the C-Safe Market Assistance Programme has provided milled sorghum to the working poor via subsidised sale through small retailers in targeted neighbourhoods in selected urban areas in Zimbabwe. It has used 'desirability' and geography as the means to target beneficiaries, e.g. self-selection, as those with sufficient income prefer milled corn where available.

Support to National Strategic Food Reserves

These programmes typically entail the transfer of grain by donors to a national strategic grain reserve or food security reserve. The main objectives of increasing national strategic grain reserves are to ensure stocks are sufficient for large-scale direct distribution in the event of an acute food crisis. This includes lending to humanitarian agencies to expedite response, and enabling the government to manipulate food markets by injecting food commodities into those markets.

This type of intervention is suited to an urban population that still has access to income to purchase food. This intervention works through national grain storage and distributions systems, thereby building capacity rather than setting up a parallel system.

Ethiopia's Emergency Food Security Reserve (EFSR) holds more than 400,000 MT of food as a reserve for distribution by the Government or to lend to humanitarian agencies in case of crisis. Via a parastatal organisation, the Government of Ethiopia began urban distributions of grain in 2008 as price inflation outstripped the capacity of poor urban dwellers to purchase staple grains.

¹Food and Nutrition Technical Assistance II Project (FANTA- 2). Emergencies in Urban Settings: A Technical Review of Food-based Program Options. Washington, DC: FANTA-2, Academy for Educational Development, 2008. Available at www.fantaproject.org

²A fully or partially state-owned corporation or government agency.