VoxEU Column Development Migration Poverty and Income Inequality

Research on humanitarian social protection is not only possible, but desperately needed

Rigorous research in humanitarian emergencies is not only feasible but also necessary to determine what constitutes effective assistance in these settings. This column introduces a Special Issue of the Journal of Development Studies which demonstrates that research establishing causal effects is vital for the design of efficient and effective social protection in settings of fragility and displacement. 

Rigorous research in humanitarian emergencies is not only feasible but also necessary to determine what constitutes effective assistance in these settings (Puri et al. 2016). A newly published Special Issue in the Journal of Development Studies demonstrates that research establishing causal effects is vital for the design of efficient and effective social protection in settings of fragility and displacement. Twenty researchers contributed seven studies examining social protection in humanitarian contexts in Ecuador, Iraq, Lebanon, Mali, Niger, and Yemen. The authors analyse various delivery modalities, such as cash transfers, vouchers, food schooling, and public distribution mechanisms. Using multiple methodologies, the studies examine the implications of several targeting choices as well as impacts across different outcomes and populations. The studies in the Special Issue confirm that rigorous, quantitative research is both possible and necessary in humanitarian settings.

Figure 1

What have we learned?

The Special Issue identifies several concrete ways to boost the impact of humanitarian social protection in fragile contexts or settings of forced displacement and migration. 

Beware of unintended consequences: The design and implementation of social protection responses in such contexts need to be carefully planned and it is important to be cautious of unintended consequences. The evaluation of Aurino et al. (in press) in Mali sounds a cautionary tale. During a challenging period, general food distribution led to declines in school attendance, primarily among boys. By contrast, school feeding led to increases in school enrolment and educational attainment – offsetting the cost of education and child labour.

Address root causes: Similarly, a study by Brück et al. (in press) of a food aid intervention in Niger finds that the provision of food aid alone did not improve child health outcomes because it did not move families out of adverse child development contexts. However, if food aid is combined with development support such as a livelihood intervention, children benefit strongly from this. Thinking about humanitarian assistance alone, without addressing the developmental root causes of a crisis at the household and village levels, is not good enough.

Outcomes are a product of design: Social protection programmes may enable households to expand investments in small enterprises and productive activities. The types of investments made, however, will depend on the design of the programme. An evaluation conducted in rural Yemen (Schwab in press), compared the impacts of direct cash and food transfers. Food transfers reduced food insecurity and enabled households to invest more in riskier cash crops. Cash transfers, in contrast, enabled households to make investments that require liquidity, including livestock. These findings matter, because household production can serve as a buffer during periods of duress.

Pause before reforming: Reforms to existing social protection programmes need to be considered carefully. Krishnan et al. (in press) study the implications of reforming Iraq’s universal food ration system, finding that reform of this system is not only politically challenging but will also have important welfare implications. Removal of these subsidies can be compensated for through direct transfers. While this reform would result in a net cost-saving, substantial transfers would be needed, especially for the poorest.

Consider different targeting methods: Preferences for a specific targeting method are unhelpful in humanitarian settings. A comparison across targeting methods in Niger (Schnitzer in press) found that proxy-means tests, household economic analysis, and geographical targeting all reach distinct groups of households. The appropriate targeting method depends on programme objectives. In a highly complex and dynamic setting, combinations of targeting methods should be considered to enhance performance.

Cash alone is not a silver bullet: The impacts of cash-based programming depend on the availability of other services. In Lebanon (de Hoop et al. in press), cash transfers enabled displaced Syrian children who were enrolled in public school to commute to school by bus. As a result, their school attendance improved. However, crowded schools due to high numbers of refugees may have dampened the effects of these cash transfers on school enrolment. Where possible, it is beneficial to coordinate and build synergies between social protection and other programmes and policies that address quality and quantity of services, for instance in the area of health and education.

Know your audience: Social protection may affect social cohesion in displacement contexts. In Ecuador (Valli et al. in press), economic transfers increased social cohesion, however these impacts were driven primarily by Colombian refugees, rather than the host population. These findings indicate that the effects of social protection should not be expected to be equal across refugee and host country populations. Careful messaging to host communities and legitimate efforts to co-target them appeared to be critical for the success of programming in this setting. 

Investing more, while making the best of what’s available

Despite these emerging insights from novel rigorous research, there remains an urgent need to expand research on social protection in humanitarian settings. Improving the efficiency of social protection has enormous potential to increase the wellbeing of displaced and host populations and, ultimately, to save lives. The choice and value of the social protection instrument, its targeting, coordination across services, and consideration of unintended impacts are all determinants of success. Investments must not only increase, but they must also be put to the best possible use through careful design and implementation based on thorough evidence. While more humanitarian aid is welcome and necessary, we should also strive to achieve the maximum impact for each dollar spent (Alda and Cuesta 2019). 

The Special Issue concludes that increasing funding for social protection in emergencies and increasing the efficiency of resources are not antipodes. Nor is thorough research in such settings – in fact, they are all desperately needed.


Aurino, E, J-P Tranchant, A S Diallo and A Gelli (in press). “School feeding or general food distribution? Quasi-experimental evidence on the educational impacts of emergency food assistance during conflict in Mali”, Journal of Development Studies.  

Brück, T, J Cuesta, J de Hoop, U Gentilini and A Peterman (in press), “Social Protection in Contexts of Fragility and Forced Displacement: Introduction to a Special Issue”, Special Issue of the Journal of Development Studies.

Brück, T, O M Días Botía, N T N Ferguson, J Ouédraogo and Z Ziegelhöfer (in press), “Assets for Alimentation? The Nutritional Impact of Assets-Based Programming in Niger”, Journal of Development Studies.

de Hoop, J, M Morey and D Seidenfeld (in press), “No lost generation: Supporting the school participation of displaced Syrian children in Lebanon”, Journal of Development Studies. 

Krishnan, N, S Olivieri and R Ramadan (in press), “Estimating the welfare cost of reforming the Iraq public distribution system: A mixed demand approach”, Journal of Development Studies.

Schnitzer, P (in press), “How to target households in adaptive social protection systems? Evidence from humanitarian and development approaches in Niger”, Journal of Development Studies.

Schwab, B (in press), “Comparing the productive effects of cash and food transfers in a crisis setting: Evidence from a randomized experiment in Yemen”, Journal of Development Studies.

Valli, E, A Peterman and M Hidrobo (in press), “Economic transfers and social cohesion in a refugee-hosting setting”, Journal of Development Studies.

525 Reads