Chloé de Soye and Cynthia Saghir
Drawing on field insights and research, this blog examines how the opaque Proxy Means Test (PMT) in Lebanon’s Cash and Voucher Assistance (CVA) system fuels misinformation, mistrust, and social tensions across both Lebanese and Syrian communities. It argues that targeting is not just a technical exercise and that over-reliance on opaque algorithms can be ineffective. This blog shows why clearer approaches are essential. It also highlights how promoting community solidarity can provide a foundation for universal social protection.
Opacity of the Proxy Means Test (PMT)
When analysing data for our study, The Role of Misinformation on Cash and Voucher Assistance for Social Cohesion in Lebanon, prepared by Key Aid Consulting for the CAMEALEON Consoritum,[1] we initially expected the main findings to focus on false rumours circulating on social media – an analysis that complemented and extended beyond traditional Focus Group Discussions (FGDs). However, the key finding turned out to be different. Echoing conclusions from a 2017 qualitative study on targeting refugee households for Cash and Voucher Assistance (CVA) in Lebanon, our research shows that the opacity of the Proxy Means Test (PMT) targeting approach[2] is the main driver of misinformation.[3]
Participants in both Syrian and Lebanese FGDs reacted strongly when asked about CVA targeting, expressing emotions ranging from anger to frustration across all regions visited.[4] Both CVA recipients and non-recipients – Lebanese and Syrian alike – viewed targeting as inconsistent, opaque, and seemingly random or luck-based.
These long-standing beliefs around targeting shape how communities perceive fairness and equity in aid distribution. Our study shows that they have fuelled tensions, weakened social cohesion, and deepened mistrust toward international aid organisations.
Is targeting in humanitarian aid something to be left to opaque algorithms?
This question might sound familiar. In 2018, Nicholas Freeland asked it in his blog for Development Pathways, calling on humanitarian agencies to wake up on the question of PMT in light of the principles of “do no harm”, transparency and accountability (including the case of Lebanon). Today, there is a broader recognition of the PMT’s limitations, in Lebanon and globally – but too often this recognition remains only on paper.
For example, the 2022 Evaluation of UNHCR-WFP Joint Action for Multipurpose Cash Assistance in Lebanon highlights that, with around 90 per cent of refugee households classified as “severely vulnerable”, the PMT targeting system struggles to distinguish meaningfully between them. The resulting rankings may appear objective but hold little practical or ethical significance for affected communities, making aid allocation decisions feel arbitrary and unfair.
Mercy Corps Lebanon recommends adopting a targeting approach that can be easily understood by communities, balancing the imperative of assisting the most vulnerable with the “do no harm” principle. Ground Truth Solutions (GTS) has repeatedly shown that communities believe aid often doesn’t reach those who need it most – such as older people, persons living with disabilities and female-headed households. They underline that commensurate impartiality earns more acceptance than decisions based on extractive surveys and vulnerability matrices. GTS also points out that even the best-intentioned targeting approaches lose effectiveness when aid is redistributed informally, calling into question the time and resources invested. People may indeed prefer receiving less if it means more people can benefit. Solidarity shouldn’t be underestimated.
Targeting: a complex technical and political exercise
Targeting is often presented as a complex technical exercise, and it is. This is especially true in protracted crises like Lebanon, where CVA is delivered by both humanitarian and development actors, including government-led social assistance. The paper Key Considerations for Targeting Social Assistance in Situations of Protracted Crises offers a useful analysis of the pros and cons of targeting approaches based on speed, cost, and accuracy. Yet it also warns that a narrow focus on the technicalities of targeting often leaves political and social considerations unaddressed.
As Mahamane Mourtala, Migration and Humanitarian Emergency Programme Officer at the NGO Karkara (Niger), put it bluntly: “A programme that lacks good targeting is doomed to fail.”[5] This is far from easy, especially as aid funding collapses.
To address these complexities, UNHCR and WFP have even created a Joint Targeting Hub. Their evidence shows that mixed methods targeting approaches can significantly improve the accuracy, inclusion, and acceptance of humanitarian aid. Similarly, the Key Considerations for Targeting Social Assistance in Situations of Protracted Crises paper recommends layering methods to balance strengths, reduce errors, and increase effectiveness. Our study in Lebanon argues that current improvement efforts focus too heavily on economic formulae – minimising exclusion and inclusion errors, rather than on a two-way communication between CVA actors and communities. Communication to affected populations on what is being done isn’t enough to improve buy-in and trust from the Syrian and Lebanese communities.
Whatever methods are used, aid actors must listen to affected populations and act accordingly. This requires flexibility, adaptation, and humility. Behavioural science illustrates why this can be difficult through the “sunk cost bias”: the more resources we’ve invested (time, effort, budget), the more likely we are to focus on those past costs rather than future gains, resulting in poorer decisions.
It is complicated to write about targeting in the current context. Global Humanitarian Overview (GHO) 2026 formalises “triage” as standard practice, marking a fundamental break from its original purpose: to present a comprehensive picture of humanitarian needs. With an appeal of just $23 billion, it effectively concedes that two-thirds of people in need will receive no assistance. As funding collapses, there is a real risk that less time and fewer human resources will be devoted to two-way communication with affected communities, and that triage decisions will increasingly rely on artificial intelligence (AI). Yet AI has real potential to responsibly automate administrative tasks to free programme staff for direct humanitarian engagement.
Towards social protection and universal coverage in Lebanon
Since 2022, significant strides have been made in Lebanon in advancing social protection, particularly in social assistance. These include the expansion of social assistance to cover 160,000 additional Lebanese households under the AMAN programme, the launch of the National Social Protection Strategy (NSPS), and the establishment of the National Disability Allowance (NDA). While these programmes have mostly been funded by international donors and loans, the Government of Lebanon (GoL) has recently allocated a budget line to the NDA to support 30,000 additional Lebanese persons with disabilities (PWDs), thereby highlighting the GoL’s fiscal commitment and shift from dependence to national ownership.
However, these initiatives primarily target Lebanese nationals and focus on expanding social assistance that is often cited as exclusionary of the most vulnerable, rather than advancing universal social protection. The fragmentation of assistance by nationality also reinforces misperceptions, particularly the belief that aid favours Syrians. This fuels social tensions and, in some cases, collective action against Syrians. By contrast, the NDA is universal, covering persons with disabilities of all nationalities. The identity of recipients of the NDA did not raise concerns along nationality lines. To enhance both trust and social cohesion, it is essential to establish a more integrated system that includes people of all nationalities. The buy-in of relevant government bodies and donors is a prerequisite. The new GoL’s Minister of Social Affairs, Haneen Sayed, provides a unique momentum with a renewed attention to social policy.
The question remains: how can universal programmes be developed and expanded, and how can international organisations support their implementation?
The GoL’s fiscal, structural and programmatic commitment to the NDA challenges the notion that a handover of humanitarian programmes to government-led social protection is not feasible or realistic. Perhaps it is not only about transitioning programmes, but it is also about choosing programmes that are operational and realistic for a government’s budget – beyond the myth that the fiscal space doesn’t exist. As evidenced in this recent publication from Development Pathways, introducing schemes gradually with a minimal yearly increase is within the capacity of most countries to finance – low-income countries included. Programmes targeted only at the poorest often shrink due to their unpopularity among taxpayers who are excluded from them. Universal, lifecycle social security schemes, by contrast, tend to have broad public support because everyone benefits from them at different stages of life. By strengthening the social contract, such schemes can create a virtuous cycle of investment in quality public services, including social security.
The NDA itself was established through the persistent efforts of a coalition of Organisations of Persons with Disabilities (OPDs), supported by the International Labour Organisation (ILO), UNICEF, as well as the Ministry of Social Affairs (MoSA), and designed by Development Pathways. This coalition – driven by advocacy for the rights of persons with disabilities and community solidarity – represents a success story in advancing universal programmes in Lebanon.
Promoting community solidarity as the foundation for a right-based, universal social protection system is a bottom-up approach that deserves greater emphasis. Community solidarity goes beyond acknowledging local dynamics – it can be leveraged to advance the right to social protection in a country where people have lost trust in the state. This resonates with Tahira Shariff Mohamed’s reflections on how to build on solidarity practices and systems instead of undermining them. Collaborating with high-profile groups such as OPDs, labour unions, and women’s rights groups, alongside thoughtful budget restructuring, can help pave an affordable and inclusive pathway toward universal social protection.
Chloé de Soye was Key Aid Consulting’s team leader for the study on misinformation. She is an independent consultant with 15+ years of experience who specialises in cash transfers, shock-responsive social protection, conflict sensitivity, and Monitoring, Evaluation, Accountability and Learning (MEAL). She led DG ECHO’s HQ policy work on cash transfers for four years, and previously held policy and knowledge-sharing roles in Brussels. She worked for 2+years for local organisations in Peru and Togo. Chloé is an experienced team leader, project manager and facilitator, and is fluent in French, English and Spanish.
Cynthia Saghir is the Social Protection Policy Advisor for CAMEALEON and Oxfam in Lebanon, works closely with the team to lead advocacy and stakeholder engagement efforts that influence Lebanon’s social protection systems. She supports research development, contributes to reports and policy briefs, and helps share findings through events and multimedia content, and provides her expertise in social protection and, more specifically, social assistance in Lebanon to accomplish CAMEALEON’s influencing and research objectives.
[1] The CAMEALEON Consortium engaged Key Aid Consulting for this study.
[2] A targeting method used to estimate households’ socio-economic vulnerability (like in Lebanon) or consumption levels indirectly by assessing observable characteristics. PMT uses a statistical model that requires a representative survey with information on expenditure (proxy for income), food security indicators and a variety of demographic and socioeconomic characteristics as well as a registration database containing relevant variables.
[3] The study identifies a total of eight significant beliefs about CVA. The methodology and areas covered are explained in the study. Targeting falls under the theme of aid bias.
[4] Key Aid Consulting conducted 12 FGDs across Beirut, Mount Lebanon, Bekaa and the North regions totalling 150 participants.
[5] Learning Event Series #2: “Using CVA in Contexts of Operational Challenges in West and Central Africa“, CALP, 8th October 2025.
Disclaimer:
This publication was co-funded by the European Union and the Norwegian Ministry of Foreign Affairs (NMFA). Its contents are the sole responsibility of the authors and Development Pathways, and do not necessarily reflect the views of the European Union or the NMFA.