Cash transfers: learning from the EU programme in Turkey

Following concerns over the targeting method used to select recipients of cash assistance in Turkey, other possible methods of targeting were tested in order to compare their relative advantages.

The EU’s Emergency Social Safety Net (ESSN) programme was introduced in Turkey in November 2016 to provide cash transfers to improve living standards for refugees who do not live in camps. It supports around 1.8 million refugees with a monthly average payment of 155 Turkish Lira (currently around 10.5 euros) per family member and quarterly top-ups depending on family size. The ESSN is the largest humanitarian programme in the history of the EU and is currently planned to continue until early 2023.[1]

Various targeting mechanisms are used for providing cash assistance to refugees globally, including; blanket targeting (that is, provided to all), geographic targeting, community-based targeting, self-targeting (where the programme is designed in such a way that only those who need assistance will choose to participate), proxy means testing (using observable characteristics of the household or its members to estimate their income when other income data are unavailable or unreliable), categorical targeting (selecting individuals belonging to a certain category of people) and using a scorecard approach. Among the humanitarian cash transfer programmes funded by the EU, proxy means testing is used in cash assistance programmes in Iraq and Lebanon, categorical targeting is used in Turkey, and something approaching a blanket targeting approach is used in Greece.

Since its launch in 2016, the ESSN programme has used categorical targeting based on demographic criteria. To be considered for ESSN assistance, household members must be registered with either international or temporary protection status and have their address registered with the Directorate of Population and Citizenship Affairs. From May 2017, to be eligible for the assistance a household had to satisfy at least one of the following demographic criteria: (i) having at least four children, (ii) having a  dependency ratio of at least 1.5, (iii) having at least one disabled individual, (iv) having a single parent or elderly head of household and (v) being a single-female household. When a household fulfils at least one of these criteria, all members of the household receive a per-person benefit, which is loaded onto a single bank card for the household. With these demographic criteria in place, the ESSN’s coverage has remained below 50% of the refugee applicants for the programme. 

Problems with targeting in a humanitarian setting

Initially, the refugee population in Turkey was generally poor with fairly uniform levels of assets and welfare. Country-wide baseline data collected in May 2017 showed that 80% of the ESSN applicant refugee population had per capita expenditure levels below the minimum expenditure basket (a poverty line determined by programme implementers).[2] The welfare levels of households were also fragile and unstable over time.

This high incidence of poverty, similar initial welfare levels and a high degree of flux presented significant challenges to targeting. The ESSN Mid-term Review[3] showed that while the targeting of the ESSN was slightly prioritising poorer families (pro-poor) in May 2017, with a quarter of the benefits reaching the bottom quintile of the refugee applicant population, by December 2018 the targeting was almost uniform across the quintiles. The evaluation also reported criticisms by refugees, including ESSN beneficiaries, about the targeting criteria. Among the 63 refugees who attended the focus group discussions conducted for the evaluation, 52 of them stated that the transfers needed to cover all Syrian refugees regardless of eligibility criteria. Frustration with the criteria continued through the following years, as evidenced by public comments such as these, in 2019 and 2020, on the ESSN Facebook page:

When will you accept four-member families? Others are getting the aid and we do not. Having two children does not mean we do not have rent and bills to pay. We are not able to cover these expenses. It is not fair.

All families should be beneficiaries, regardless of the number of children. These criteria must be abolished.

Alternative targeting options

UNHCR and WFP state that targeting may not be appropriate in some cases[4] such as (i) in the immediate aftermath of a crisis that affects most of the population, where needs are very high and the population is more homogeneous (when targeting may create additional tensions), (ii) in a situation where targeting is not methodologically or practically feasible due to limitations in capacity or time, or when there is a lack of available data or access restrictions, or (iii) in a situation where the cost of targeting is higher than providing assistance to everyone.

The ESSN mid-term evaluation by WFP simulated the changes in poverty rates under different targeting scenarios. These simulations suggested that a more universal approach to targeting could have marginally improved the refugee poverty headcount (percentage of those under the poverty line) while being easier to implement – and would have been perceived as more equitable. In the simulated data, the poverty rate of the overall refugee population prior to receiving cash assistance was calculated as 76%. Distributing the transfer to households using the initial ESSN demographic targeting criteria was simulated to reduce the poverty rate to 69%. Distributing the transfer to all refugee applicants while keeping the total cost the same (that is, by reducing the amount of cash transferred) resulted in a simulated poverty rate of 70%, whereas distributing the transfer to all refugees was found to result in a poverty rate of 71%.

The Mid-Term Evaluation of the Facility for Refugees in Turkey also tested alternative targeting scenarios, based on targeting the cash transfer universally to (i) all children, (ii) the elderly, (iii) women living alone and (iv) disabled individuals with a medical report. The report found that the impact of such a universal transfer (in terms of reducing poverty headcount) would have been much enhanced. The simulated change in coverage based on a simple revision of the eligibility criteria results in a tremendous increase in coverage from 48% of the refugee population living in a household receiving transfers to 91%.[5] The total cash transfer cost increased by 14% in this scenario, assuming that all eligible families would apply for the grant. When the top quintile (20%) of refugee applicants was excluded from the programme in this simulation, this approach was completely cost-neutral when compared with the current version of the targeting criteria. This ‘almost universal’ approach – excluding only the top quintile and including all children, elderly and disabled in the programme – would result in a lower poverty headcount rate of 47%, compared with the original approach which resulted in a 50% poverty headcount rate, while the coverage increased from 48% to 78%. 

The ESSN programme offers important lessons for donors and programme implementers. What we learned from the ESSN targeting experience is that a blanket approach that covers the majority of refugee households with a per capita grant for dependents (children, elderly, disabled) may have performed better as a targeting strategy. The coverage of refugees would have been higher, and the poverty headcount ratio would have been reduced more than with the current version of targeting, at a similar (or slightly higher) cost.  Hence targeting efforts using PMT scores should focus on ‘targeting out’ the top of the distribution and providing an almost universal grant for the dependent family members for the rest of the distribution, rather than trying to target the poorest refugee households.

Given the experience in Turkey, adopting such an almost universal approach to targeting in other countries may be operationally less costly and socially more acceptable within the refugee community. This approach would reduce the high stakes associated with the programme, and reduce social tensions between the beneficiary and non-beneficiary groups. As the benefit level per refugee household would be lower (and the working-age members of the households would be excluded from the grant), it is likely that the social tensions with those in the host community who benefit from social assistance would also be avoided.   


Meltem Aran

Director, Development Analytics


Nazli Aktakke

Senior Social Policy Researcher, Development Analytics


Hazal Colak

Data Scientist for Social Policy, Development Analytics


Gokce Baykal

Independent Researcher


[1] The ESSN was managed by the World Food Programme until 2020 and subsequently by IFRC, supported by the Ministry of Family and Social Services (MoFSS), the Turkish Red Crescent and Halkbank. See DG ECHO ‘The Emergency Social Safety Net (ESSN): Offering a lifeline to vulnerable refugees in Turkey’ Most recently, ‘C-ESSN’ took over a share of the caseload in July 2021. As of December 2021, the ESSN caseload was around 1.5 million individuals, while C-ESSN covered 363,464 individuals.

[2]  World Food Programme (2018) ‘Evaluation of the DG ECHO funded Emergency Social Safety Net (ESSN) in Turkey, November 2016–February 2018 Volume 1: Final Evaluation Report – Turkey’

[3] World Food Programme (2020) ‘ESSN Mid-Term Review 2018/2019, February 2020 – Turkey’

[4] UNHCR and WFP (2020) Joint Guidance: Targeting of Assistance to Meet Basic Needs

[5] EU Directorate-General for Neighbourhood and Enlargement Negotiations (2021) ‘Strategic Mid-Term Evaluation of the Facility for Refugees in Turkey (2016-2019/20)’


Avis de non responsabilité
Les avis contenus dans RMF ne reflètent pas forcément les vues de la rédaction ou du Centre d’Études sur les Réfugiés.
Droits d’auteur
RMF est une publication en libre accès (« Open Access »). Vous êtes libres de lire, télécharger, copier, distribuer et imprimer le texte complet des articles de RMF, de même que publier les liens vers ces articles, à condition que l’utilisation de ces articles ne serve aucune fin commerciale et que l’auteur ainsi que la revue RMF soient mentionnés. Tous les articles publiés dans les versions en ligne et imprimée de RMF, ainsi que la revue RMF en elle-même, font l’objet d’une licence Attribution - Pas d’Utilisation Commerciale - Pas de Modification (CC BY-NC-ND) de Creative Commons. Voir pour plus de détails.