While recent years have seen a shift towards alternative food assistance instruments in emergency and transition situations, hard evidence on their performance is limited. To help fill this gap, the German Agency for International Cooperation (GIZ), on behalf of the German Federal Ministry for Economic Cooperation and Development (BMZ) and in collaboration with researchers from the University of Juba, South Sudan, conducted a comparative study on the efficiency, effectiveness and impact of food for work (FFW) and cash for work (CFW) interventions. Analysis of the data reveals that, irrespective of differences in livelihoods among the respondents, CFW performs better than FFW on indicators such as cost-efficiency, expenditure autonomy, dietary diversification, quality of works and participants preferences.
Project setting and hypothesis: does the socio-economic context matter?
The evaluation took place in the context of a BMZfunded Transitional Aid Project in Magwi and Morobo counties in South Sudan, with an FFW component and a subsequent CFW component. The main focus of the project is livelihood support for returning refugees and local inhabitants. For the purpose of the study, two rounds of household surveys with a total sample size of 332 respondents were conducted: a first round in autumn 2010, to collect baseline data, and a second round in spring 2011.
One question about the performance of CFW and FFW relates to their feasibility and suitability in different socioeconomic contexts. Since cash transfers are only successful if the money can be used to purchase goods and services, it is often assumed that CFW is less suitable in very early post-conflict scenarios, where people are still displaced, markets are disrupted and infrastructure is damaged. A livelihood analysis conducted as part of the study led to the initial hypothesis that FFW was more appropriate in Magwi than in Morobo. While food consumption scores in both Magwi and Morobo counties were very poor, with the vast majority of respondents reporting having suffered food shortages in the past, the overall situation in Morobo appeared more consolidated than in Magwi, and a higher number of Morobo households had been settled or returned for longer. Families in Morobo were bigger on average, and had more land and assets. Moreover, markets in Morobo appeared to be better supplied than those in Magwi: 47% of interviewees in Morobo reported occasional shortages of basic food items on local markets, compared to 73% in Magwi. Given that food needs were likely to be greater in Magwi and cash of less use because markets were not functioning properly, FFW was expected to perform better.
The overall aim of the research was to compare the effectiveness, efficiency and impact of the CFW and FFW projects in Magwi and Morobo. Effectiveness was defined as the extent to which the interventions immediate objectives were achieved, measured by the timeliness of payments and the targeting of individuals most at risk. Efficiency was measured by the relation of project inputs and outputs (i.e. cost-efficiency). Finally, impact relates to the wider effects of the intervention, such as the contribution of CFW/FFW to household income and the impact on food consumption, local markets and the work mentality of participants.
Effectiveness: beneficiary targeting and timeliness of payments
Targeting the poorest section of the community through a mix of self-targeting and administrative criteria was relatively successful: 72% of participants in Magwi and 90% in Morobo assessed their own livelihood situation as being either destitute or poor. In both counties, around 60% of the participants cultivated 13 feddan, while the share of non-participants with land of over three feddan was significantly higher than that of participants. One feddan = 60 x 70m On the other hand, the share of nonparticipants cultivating less than one feddan was higher than among participants. What is more, participants in Magwi indicated owning more household assets and livestock than non-participants. There was also some evidence that working group chairs played an important role in selecting participants, especially in Magwi, where the supply of workers was higher than the demand for them.
According to the results of the first survey, around half of the FFW participants did not receive payments on time, partly due to difficulties in procuring food items from Uganda. CFW payments were generally received on time or with only minor delays.
Efficiency: cost, value and time
Overall, the project costs for FFW rations (including procurement, transport and handling) were some 20% (SDG 3) below the local market value of the FFW rations in 2010, i.e. less than the money needed to buy the same food basket on local markets. Even so, FFW cannot definitely be described as cost-efficient since the costs for project staff could not be included in the calculation, and FFW rations were of poorer quality than food sold on the markets. Due to the low quality of the food distributed, FFW rations fell short of the value of CFW payments (which were based on the market value of food baskets). This meant that CFW recipients could buy more and/or better food.
In terms of time expenditure, FFW appears to be less attractive since managing logistics, coordinating food transport from Uganda and distributing rations absorbs a large amount of time. Project staff report that handling cash is much less demanding.
Impact: household income, markets and consumption
CFW and FFW interventions do not provide a full income, but rather a complementary source of household revenue during critical months when there is a lack of alternative income and employment opportunities. Based on income at poverty-line level (SDG 72.9 per capita per month) and an average household size of 7.5, the CFW/FFW participants could receive, on average, up to 55% of the monthly income required to reach the poverty line, if they were paid for 20 working days per month. Owing to the repeated suspension of works, on average income amounted to 41% of monthly poverty line income.
With CFW payments, beneficiaries were able to buy what they chose. While many FFW participants usually sold part of the food ration to generate cash for other expenditures (despite in many cases not achieving the full market price for their sales), the majority of CFW participants spent a major portion of their payments on buying food. CFW therefore provides greater autonomy of choice for beneficiaries.
Both FFW and CFW participants reported changes in their household food consumption habits, mainly in the form of more meals per day or more food per meal. However, only CFW participants stated that they also ate different types of food, an indicator of increased food diversity.
The research was not able to provide evidence that CFW or FFW had had any significant impact on the local food market since the volume of food rations and cash payments made under the programme was too small to affect aggregate supply and demand. Also, impacts on the availability of some seasonally sold food items could not be measured by the surveys. It is likely that local markets would respond to increased demand should CFW, and consequently the availability of cash, be expanded.
In terms of participants preferences and attitude to work, project staff and beneficiaries alike report that CFW has a positive impact on work mentality and on the quality of the work done. While communities tend to view FFW as a charitable operation, CFW is considered an employment opportunity. This positive work attitude also reflects a preference for CFW revealed by the surveys. Although this preference was observed when FFW alone was being offered by the project, far more community members expressed an interest in participating when CFW activities started. Interestingly, a higher percentage of female than male participants declared a preference for FFW or a combination of CFW and FFW.
The main message from the study is clear and precise: to shift, as much and as rapidly as possible, from FFW to CFW in Morobo and Magwi counties. Even in Magwi, despite the difficult socio-economic conditions, the wider application of CFW is justified. The strong preference among participants for cash calls for an expansion of activities in and beyond these two counties, the identification of additional resources, public works and alternative payment modalities and better targeting approaches in view of the high demand for CFW. Early and repeated analyses of local food markets, cost-efficiency aspects and beneficiaries preferences may be difficult in transitional settings, but they are clearly indicated when choosing a suitable method of food assistance or expanding a specific intervention.
While the South Sudan analysis is not directly transferrable to other countries and scenarios, the following general lessons can be drawn from it. First, cash instruments should not be dismissed in the planning phase simply on the basis of preliminary assumptions about livelihoods. Rather, an early, and possibly repeated, analysis and review of food assistance instruments applied in transition situations should be the standard approach. Such an assessment will have to consider current and changing needs, the market situation, cost-efficiency, the institutional infrastructure, management capacity and beneficiaries preferences. If changes are observed, project strategies and instruments will need to be adjusted accordingly. Table 1, informed by the results of the study and other research conducted by the authors, lists the issues that should be considered when deciding on suitable food assistance instruments and payment modalities.
The second lesson from Magwi and Morobo is that participants preferences are just as important as an analysis of markets and infrastructure. The analysis shows that beneficiaries satisfaction contributes to the success and quality of the work done. In this regard, it is not advisable to implement a combination of FFW and CFW, or parallel implementation in different sites, if one option is clearly favoured over the other. Given the growing consensus that interventions in early transition scenarios should contribute to capacity-building and coping, insights about participants expenditure autonomy, preferences and work mentality indicate important advantages of CFW over FFW and probably not just in South Sudan. The risk with FFW is not only that it might hamper and distort local food markets, but that it might also make it harder for beneficiaries to achieve independence and resilience. Experience in Magwi and Morobo shows that following community preferences will attract more participants. Consequently, whichever instrument is chosen, targeting has to be adjusted to local demands and needs, as well as to gender differences.
Manfred Metz and Anastasia Guretskaya are consultants. Melha Biel and Henry A. Kenyi are members of the University of Juba.