Postscript: Dealing with urban emergency: lessons from Oxfam’s EFSL activities in three cities

By Ian MacAuslan and Laura Phelps

Ian MacAuslan leads Oxford Policy Management (OPM)’s education, early childhood development and labour portfolio and is a senior consultant in social protection. He works on programme strategy, design, monitoring and evaluation, specialising in qualitative research on access to education, social protection, child labour, food security, and urban issues.

Laura Phelps is a public health nutritionist with over 14 years of experience in food security and livelihoods, in both emergency and recovery phases. For the last 11 years she has worked for Oxfam, and since 2010 she has been the cross sector urban humanitarian lead for programme and policy development. Laura has recently moved to the Norwegian Refugee Council where she will be the global Urban Displacement Expert focusing on developing NRC as a leading urban agency for urban displacement.

The authors acknowledge the contributions of Oxfam staff, particularly the team in Nairobi and a special mention for Sumananjali Mohanty, Oxfam Kenya.

This article considers the findings from the assessments of three Oxfam urban emergency food security and livelihoods (EFSL) programmes in Nairobi (Kenya), Port-au-Prince (Haiti) and Gaza¹ (insights into these programmes are provided in three separate field articles in this issue of Field Exchange²). The purpose of the assessments was to explore:

The rooftops of the Nairobi slums, where up to 60 per cent of the city’s inhabitants live

• Appropriateness, timeliness and impact: whether programmes were well designed for the emergency and were initiated in a timely manner such that they had a positive impact.
• Targeting: whether programmes were able to select those most in need of assistance in a timely and cost effective manner.

The article describes some characteristics of urban vulnerability, and then considers how well these programmes have been able to perform, before drawing some lessons for organisations seeking to work more on urban emergency response.

The methodology for each study was based on a review of documents and semi-structured interviews and group discussions with programme officers, other stakeholders, and participants in the programmes.

Urban vulnerability: a growing challenge

Urban vulnerability is not very well understood but four clear points are worth noting.

First, urban populations now exceed rural populations globally (as shown in Figure 1) and are growing very quickly. The slum population is also growing rapidly, though there are considerable uncertainties about how many people live in slums, and how quickly they move in and out.

Sources: World population from World Development Indicators. Urban populations are defined as those living in settlements over 1 million. Slum population from State of the World’s Cities 2011, page 32.

Second, urban vulnerability is based on exposure to risks around markets, health and sanitation, and social and political relations. This is typically more complex than the risk profiles in rural areas. The market exposure means that responses working through cash are often the most appropriate. The social and political fragmentation means that targeting mechanisms cannot easily work through communities and face risks of capture by powerful actors for their own purposes.

Third, while urban poverty rates seem lower than rural poverty rates in most countries, slum poverty rates are often higher and more intense, with child mortality and health indicators particularly poor. Urban food insecurity is a very real issue, but is generally not as severe as in rural areas. Although malnutrition prevalence rates tend to be higher in rural areas, high urban population density means that there are many more malnourished individuals per unit of area in slums.

Fourth, urban areas change very rapidly. The rapid change means that it is vital to have clearly defined triggers for scaling up and scaling down emergency programmes.

How well have Oxfam’s urban EFSL programmes fared?

The three programmes described in the field articles differed substantially and responded to very different types of vulnerability:

• The Nairobi Urban Social Protection Programme (NUSPP) responded to stagnant incomes and price increases and spikes related to the drought in the Horn of Africa in 2009. This is a slow onset urban emergency. The response included cash transfers, cash for work, skills training and business grants, and advocacy to the government for scale-up.
• Oxfam’s Emergency Food Security and Livelihoods (EFSL) responded to the earth quake in Port-au-Prince in 2010. This is a rapid onset urban emergency. The response included cash for work, food and cash grants, training and in kind transfers.
• The Gaza Food Security and Livelihoods Programme (GFSL) responded to a chronic political and economic crisis, as well as the protracted impact of the Cast Lead military operation in the winter of 2008/09. This is a conflict-related urban emergency. The response included cash for work, a voucher programme, training and support to income generation.

Appropriateness, timeliness and impact

Overall, the EFSL interventions were highly appropriate, but varied in their timeliness, and therefore their programme impact.

Nairobi

The objectives of the NUSPP were appropriate to the acute stress overlain on the chronic poverty of urban informal Nairobi, and the intervention was timely. Given the extent of vulnerability in urban informal areas, it was sensible to engage with the government and others to achieve a greater scale and to aim to develop a government-led programme. There were a number of design features that could probably have been adjusted to maximise impact. For example, the targeting approach and process could have been more clearly specified and the value of some transfers could have been revised to ensure the most vulnerable households received the transfer for longer, rather than a small number of households benefitting from a large transfer.

The impact of the NUSPP was very positive. First, the cash transfer and livelihoods activities improved the food consumption of recipients, saved lives of those on anti-retrovirals, helped them reduce (though not always avoid) the use of negative coping strategies (such as prostitution, crime, and removing children from school), helped recipients pay off debt and helped some recipients start or restart businesses, some of which are generating positive returns. The impact of the business training and grants is less clear, but seem positive on a small-scale, though these are unlikely to be retained at scale. Second, the government adopted the cash transfer aspects of the programme and is in the process of scaling these up, thereby greatly enhancing Oxfam’s impact.

Port-au-Prince

The Port-au-Prince EFSL objectives were in general appropriate to the post-earthquake context, but the interventions appeared to be based more on pragmatism than on a comprehensive livelihood analysis. Deeper analysis to start would have helped to allocate resources more efficiently, but there was also a compulsion to respond quickly. Some aspects of value for money - such as the choice of cash transfer provider - could have been improved.

The EFSL activities had different positive impacts individually. Taken together, however, it is not clear that they were able to prevent the serious negative consequences of the earthquake as most recipients continued to cut essential expenditures from their budgets. This should be seen in the context of the scale of the disaster. Positive impacts included:

• The short-term guaranteed employment programme helped households repay debt and cover basic needs.
• Community canteens provided food to the most vulnerable and this increased the number of meals they consumed per day.
• Cash transfer programmes increased their credit worthiness and financial inclusion. The impact was constrained by delayed implementation.

Gaza

The GFSL has evolved over time in relation to the changing context in Gaza. This includes support to a long-term, chronic humanitarian situation and an emergency response following Cast Lead, the Israeli military offensive in 2008-09. The programme operates in a complex environment with limited opportunities for diversified funding and with restrictions imposed on engagement with the local authorities. Activities could have been made more urban specific for sustainability and the efficiency of implementation improved.

The GFSL had broadly positive impacts on recipients without significant negative impacts.

Positive impacts include:

• The cash for work programme helped participants to repay debt and cover basic household needs, and improved their creditworthiness, but did not lead to further investment.
• The programme developed the social capital of participants through a greater feeling of selfesteem and meeting new people.

Cash transfer beneficiaries engage in food vending in the slums

Targeting

Overall, the programmes targeted reasonably well, but were let down by high levels of exclusion error in Nairobi and Port au Prince as well as unsystematic approaches. Targeting indicators and methodology, as well as trigger indicators for entry and exit, require more field based research and design.

Nairobi

The NUSPP sought to target the most vulnerable households in urban informal Nairobi. The targeting process for the cash transfers used local officials and volunteers to develop indicators of vulnerability and then to identify vulnerable households. The targeting outcomes were probably reasonable in terms of exclusion and inclusion errors, which is impressive given the challenges of urban targeting. However, the reliance on these local officials with minimal effective external verification raises some significant risks for targeting effectiveness. More experimentation with different methods going forward (including a census and scorecard-type approach) would help learn lessons for a scaled programme. The process of allocating exit activities (cash for work, skills transfer and business grants) relied too much on self-targeting and word-ofmouth, and was probably therefore not as effective.

Port-au-Prince

The Port-au-Prince EFSL targeting process was participatory and specific, using a household level scorecard. Oxfam staff spent considerable time on verification and this delay caused households to lose resources. Perhaps, however, given the scale of disaster, blanket targeting, or targeting using an indicator that included isolation (e.g. geographic distance from markets) or displacement (e.g. whether the household has been forced to move by disaster), might have used resources more effectively.

Gaza

The GFSL used a very well structured targeting analysis methodology using community rankings to target geographically and scorecards to target households. This led to a high degree of targeting effectiveness in each of the different project interventions. Nevertheless, this could have been improved through making the scorecard indicators more specific to the urban context and possibly including indicators such as isolation, displacement and low food consumption and human capital. Households who had recently fallen into poverty (for instance due to the blockade) were less likely to be targeted until they had lost their assets. It may be that assisting them before this would be more efficient. However, it is recognised that with funding limitations and large numbers of eligible recipients, this may not always be possible. Currently, proxy measures have combined with food consumption scores to refine targeting, and a complaints mechanism has been set up in 2012.

Conclusions: how can organisations engage in urban emergencies?

EFSL work in urban areas is gaining importance and will continue to grow in importance as urban populations grow and their vulnerabilities increase, particularly if food prices continue to be volatile. Oxfam has been generally effective in implementing EFSL programmes in urban areas, and these programmes seem to have a significant positive impact on a growing urban vulnerable population.

For most organisations, implementing urban programmes will require overcoming some significant internal and external constraints. These are partly due to the historical (and previously justified) balance of resources in most organisations in favour of rural areas, but also to some attitudes that urban areas are better able to cope. The assessment here suggests that these attitudes are decreasingly appropriate. Organisations should therefore devote resources to improving the share of their programming that is urban.

Addressing situations of chronic poverty (as is the case in all three case studies here) requires a much more comprehensive and a better resourced approach than is currently used. Approaches such as Oxfam’s ‘one programme approach’ are important and urban programmes should be designed to cover the same geographical areas and enable an ‘expand and contract model’ for scale up to emergency response, followed by transition back into the development programme. This would promote urban resilience.

EFSL programming in urban areas should aim to prevent households’ welfare from further deteriorating through the use of negative coping strategies, and to help households begin the process of rebuilding. However, EFSL programming cannot be expected to address underlying vulnerabilities or chronic poverty, and this is where good governance and sustainable livelihoods programming needs to compliment EFSL work. For example, safety net cash transfers have longer term social protection objectives, but these should be supported through good governance and essential services support.

We conclude with some brief specific recommendations:

• NGOs have an important role to play in terms of developing and testing appropriate responses to food insecurity and crisis, and engaging with governments and other stakeholders to help to institutionalise these responses.
• Triggers for entry and exit for urban social protection responses are vital but currently not developed³.
• Further comparisons are needed between cash transfers, poverty scorecards, extractive questionnaires and proxy means tests in urban areas.
• Cash transfers are an effective first response to urban disasters to meet immediate needs in food, water, shelter, transport and healthcare.
• Innovative technology used appropriately can be very helpful to speed the response (through digital data gathering) and make transfers (through phones or bank cards).
• As urban programming is a relatively new area with little established evidence based practice, all new programmes should be designed and funded with a strong M&E capacity, as well as a significant research component, to ensure that learning, particularly on targeting, is fed back into future scale up and replication of models developed.

For more information, contact: Ian MacAuslan, email:ian.macauslan@opml.co.uk

¹Oxfam GB Emergency food security and livelihoods urban programme evaluation. Final report. Ian MacAuslan and Laura Phelps, March 2012. The full report is available from OPM, see contacts at end of article.

²The Haiti and Nairobi field articles include updated information that will not have been considered in the OPM assessments

³For more recent work on this since this review was carried out, see the final report of a review of urban food security targeting methodology and emergency triggers, by Ian MacAuslan and Maham Farhat (July 2013). This report is summarised in this edition of Field Exchange.