humanitarian · digital · family: the rulebook was written for a world that no longer exists
no id, no aid
Digital humanitarian cash transfers exclude people without identity documents
Problem statement
Cash and voucher assistance has been one of the biggest changes in humanitarian action in the last decade: in 2019, $5.6 billion of international humanitarian assistance, 17.9% of the total, was delivered as cash or vouchers, and more and more of it is paid digitally. But digital payments require a personal account with a financial service provider and identity checks that many people cannot satisfy: the World Bank estimates that 1 billion people, 13% of the world's population, have no official proof of identity, and armed conflict makes this worse, sometimes leaving people in territories where their documents are no longer recognized as valid. KYC (Know Your Customer) rules — set globally by the Financial Action Task Force, written into national law, and applied by financial service providers — are read by most States as requiring a legally accepted ID such as a passport, birth certificate, or national identity card, which can shut out people who have none, the poorest among them. The result is a paradox: digital cash is more efficient and dignified than in-kind aid, but digitization itself can create a new exclusion barrier.
Why this matters
People who cannot use digital payments have to be offered something else — physical cash or in-kind aid — and the ICRC article says offering different options can be challenging: cash gives people more choice than in-kind aid and gives agencies more operational flexibility, paying cash by hand to large numbers of people can be logistically complicated, and setting up a new in-kind pipeline takes time. When most of a population can use digital payments, agencies choose them, and the groups without access can be left out unintentionally. The article also warns that interoperable identity systems could carry significant risks for vulnerable groups who may have very good reasons to stay anonymous and could face discrimination or worse by being identifiable, that biometrics collected by humanitarian organizations are of particular interest to States, that the data agencies hold can be valuable to parties to a conflict, and that payment data shared with financial service providers can be used by authorities for surveillance and profiling or exposed by hacks.
What’s been tried and why it hasn’t worked
Biometric identification has spread quickly among humanitarian organizations because it is seen as bringing efficiency and accountability, especially against fraud, but it creates data protection risks: a 2016 UN internal audit found that in four of five country operations reviewed, refugees were not given enough information about UNHCR's biometric programme to be properly informed. Blockchain-based "self-sovereign" identities, where the person controls who can access their identity data, are raised in the ICRC article as a possible middle ground; in a separate passage the article cites a study of humanitarian cash assistance finding that, despite interest in distributed ledger technology and digital currency, "real-life application at scale is a long way off." Sending the payment provider only a unique identifier and an amount, so the provider cannot identify the person, can work for people with no formal identity document, but it gives them no personal account and so no financial inclusion. Helping people obtain missing identity papers through legal services, and even paying for them, is another route the ICRC and other agencies use, though failure to meet KYC requirements "can be harder to resolve." Refugee ID cards issued by UNHCR, where negotiated with the relevant authorities, give access to services beyond UNHCR's own, including financial services in some cases. Mobile money works where people already use it — the ICRC makes most of its payments in Somalia that way — but in places such as South Sudan, where means of digital payment are extremely limited, small payments are still made in cash. The ICRC article's own conclusion is that the risks of digital payments can never be entirely avoided, only mitigated, and that agencies must always offer alternatives.
What would unlock progress
Privacy-preserving digital identity systems using zero-knowledge proofs or similar cryptography — verifying "this person is a registered beneficiary" without revealing identity. Offline-capable digital payment systems for settings without connectivity. Regulatory frameworks creating humanitarian exceptions to standard KYC for small-value transfers. UX design for digital illiterate users that maintains security without requiring literacy or smartphone proficiency.
Entry points for student teams
A team could prototype a zero-knowledge proof system for humanitarian aid eligibility verification, demonstrating how beneficiary status can be confirmed without transmitting personal data. Alternatively, a team could design and test an offline-capable payment token system usable by people with no prior digital experience. Cryptography, UX design, and humanitarian technology skills apply.
Genes
Structural cousins: stuck the same way, in other fields
Sources
Jo Burton (ICRC), "“Doing no harm” in the digital age: What the digitalization of cash means for humanitarian action," International Review of the Red Cross, vol. 102, no. 913, pp. 43-73 (IRRC No. 913, March 2021), accessed 2026-02-24; UNHCR Cash-based interventions, accessed 2026-09-30 go to source 1 ↗ go to source 2 ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
The identity exclusion problem may deepen as more humanitarian agencies shift to digital cash — the very efficiency gains that make digital cash attractive increase the stakes of digital exclusion. The ICRC's "doing no harm" framework explicitly flags biometric data risk in conflict settings. Distinct from `humanitarian-refugee-disability-digital-access` (which covers disability-specific digital access barriers) — this brief addresses the broader population excluded by identity documentation requirements.
Reconciliation 2026-09-30: Fact repair after reading the whole write-up against its sources, with a second reader going over every repair on the same day. The first source, the International Review of the Red Cross article, is blocked to scripts at the journal's own address (a firewall page), so it was read in full text, footnotes included, from the Internet Archive's copy of that page (https://web.archive.org/web/20240529105125/https://international-review.icrc.org/articles/doing-no-harm-digitalization-of-cash-humanitarian-action-913, fetched 2026-09-30 by both readers); its citation record was fetched from Crossref (https://api.crossref.org/works/10.1017/s1816383120000491). The second source, "UNHCR Cash-based interventions", was named with no address; UNHCR's site also blocks scripts, so its cash-based interventions page was read in full from the Internet Archive's copies of 2026-09-28 and 2023-07-01 (https://www.unhcr.org/what-we-do/respond-emergencies/cash-based-interventions). That page is a programme statement with links to more than sixty publications; the linked publications are not sources this write-up names, so nothing was taken from them. Changes: (1) The Source line credited the article to "ICRC" with a shortened title. The record and the page give the author as Jo Burton, the ICRC's institutional lead for cash and voucher assistance, writing in her own name with a note that the views are hers; the title, volume, issue and pages were copied from the record, and the address of the UNHCR page was added; the journal's link is unchanged. (2) "Cash-based assistance is now the preferred humanitarian aid modality, with UNHCR, WFP, and UNICEF collectively delivering billions in cash transfers annually" is in neither source. The article says cash and voucher assistance has been one of the biggest changes in humanitarian action in the last decade and that in 2019, $5.6 billion, or 17.9% of international humanitarian assistance, was delivered through cash or vouchers, and that funds are increasingly disbursed digitally; the sentence now says that. (3) The figure that "20-30%" of displaced people cannot satisfy identity checks is in neither source. The article gives the World Bank's estimate that 1 billion people, 13% of the world's population, have no official proof of identity, and says armed conflict makes this worse; the text now carries those figures and gives no displaced-specific share. (4) "Refugees often lack government-issued ID, SIM card registration requires ID in most countries" is stated nowhere in either source; the article's reasons why people lack proof of identity (cost of documents, literacy, gaps in national systems, conflict) are covered by the sentence above, and the SIM claim was removed. (5) "KYC regulations from financial service providers systematically exclude the most vulnerable" overstated the article, which says KYC rules are set globally by the Financial Action Task Force, transposed into national law and applied by providers, that most States read them as requiring a legally accepted ID, that the cost of documents is a barrier for the poorest, and that exclusion of those without access is "possible - albeit unintentional"; the sentence now says that, and "creates a new exclusion barrier" became "can create". (6) "Forces the most vulnerable people back to in-kind aid distribution - less dignified, less flexible, and more expensive to deliver": the article says the alternatives are cash payments or in-kind assistance, that cash increases people's choice and gives agencies greater operational flexibility than in-kind aid, that paying cash to large numbers is logistically complicated and a new in-kind pipeline takes time; it does not say in-kind aid is more expensive, and that claim was removed. (7) "In Jordan's Azraq and Zaatari camps, biometric-linked cash systems serve ~100,000 refugees effectively, but populations outside formal camps and those who avoid registration remain invisible" is in neither source and was removed; in its place is the article's statement that interoperable identity systems could carry significant risks for vulnerable groups who may have very good reasons to remain anonymous and could face discrimination or worse if identifiable. (8) "The ICRC warns that centralized biometric databases ... create surveillance and persecution risks if compromised": the article does not use "centralized" or "persecution"; it says biometrics collected by humanitarian organizations are particularly interesting for States, that humanitarian data can be valuable to parties to a conflict, that data shared with providers can be used by authorities for surveillance and profiling, and that payment data can be hacked; the sentence now says that and attributes it to the article. (9) In "What's Been Tried", five claims were in neither source and were removed: IrisGuard iris scanning in Jordan with "expensive enrollment infrastructure"; a UNHCR blockchain pilot on Stellar/USDC; that prepaid cards and vouchers bypass KYC; the ICRC's Humanitarian Token Solution with QR codes; and that no system prevents duplication fraud, works offline, protects against database compromise and works without digital literacy. The section now carries what the article reports as tried: biometrics adopted for efficiency and fraud control, with the 2016 UN internal audit finding on four of five country operations; blockchain-based "self-sovereign" identities as a possible middle ground and, from a separate passage of the article, its quoted study finding that distributed-ledger application at scale is "a long way off"; passing the provider only a unique identifier and an amount, which works without an ID but gives no personal account; legal help to obtain identity papers; UNHCR-issued refugee ID cards giving access to financial services in some cases where negotiated with authorities; and mobile money in Somalia against cash in South Sudan. (10) "The identity exclusion problem is growing" in the notes above became "may deepen", the strength the article gives it. Confirmed against the sources and left as written: the long title and display title; that cash assistance is more dignified than in-kind aid; that digital delivery is faster and more efficient; the "do no harm" framing and its flagging of biometric risk in conflict settings; and the cross-reference to the disability brief. The second reader fetched the same three records independently, found every repair supported, and tightened three sentences to the article's own strength: "the ICRC article is plain that switching is hard" became "says offering different options can be challenging" (the article: "It can be challenging to offer different options"), interoperable identity systems "carry" significant risks became "could carry", and the "long way off" finding is now marked as coming from a separate passage of the article rather than as its caveat on self-sovereign identity. Not checked: the proposals in "What Would Unlock Progress" and the entry points, which are the write-up's own suggestions rather than claims about the sources, are left as written; the removed Jordan, Stellar and Humanitarian Token claims may be sourceable elsewhere but are not sourced here. Genome tags untouched.