UNHCR Delivers $35M+ in Aid to 625,000+ Afghans via Algorand-Powered HesabPay

The Milestone, and the Question Underneath It

As the Humanitarian Aid Payments Council convened its third annual meeting in Washington, D.C. on July 15, 2026, the UN Refugee Agency (UNHCR) released a number that gives the blockchain-aid movement its strongest institutional proof point yet: it has scaled its use of reloadable cards through HesabPay¹, the Algorand-powered Afghan payments platform, to support more than 625,000² refugee returnees and over 17,500³ internally displaced people, with more than $35 million⁴ in assistance delivered. 'Blockchain-powered payment infrastructure that is locally connected, globally compliant, and fully traceable has the potential to strengthen trusted aid delivery,' UNHCR Corporate Treasurer Carmen Hett said in the Algorand Foundation⁵ release that accompanied the data.

That milestone raises the question this article investigates: how much of the humanitarian story Algorand has been telling since 2022 is real on-the-ground deployment, and how much is positioning? Three consecutive years of Council meetings, a UNHCR-sourced scale-up, a Mercy Corps pilot with published metrics, and a June 2026 report from CGAP⁶, the financial-inclusion research group housed at the World Bank, now make that question answerable against primary documents.

First, the problem the entire effort targets. Humanitarian cash and voucher assistance (CVA) — aid delivered as money or vouchers rather than in-kind goods — reached $10.6 billion in 2022, about 24 percent of all humanitarian aid. But getting funds to recipients requires two legs: a cross-border transfer that moves bulk money into the country, and local delivery to individual recipients. Both legs are broken in exactly the places need is greatest. Global banks, under anti-money-laundering compliance pressure, have been 'de-risking' — dropping correspondent relationships in high-risk jurisdictions, the mechanism by which banks in different countries hold accounts with each other to process international transfers — leaving some countries entirely cut off from international payment rails. Where correspondent banking still functions, cross-border transfer costs average 6.5 percent globally and more in fragile states, and multi-hop routing can hold transfers for days or weeks.

Kabul's Digital Wallet That Outgrew the Crisis

At the center of the Afghan deployment is Sanzar Kakar, an Afghan-American entrepreneur who started as a technology analyst at Merrill Lynch and later led investment and economic-advisory programs funded by the U.S. State Department and UK aid in Afghanistan. He founded HesabPay in Kabul in 2016 as a digital wallet, in a country where Gallup estimated that 97 percent⁷ of the population lives below the poverty line. After the regime change of August 2021, with more than 22.8 million⁸ Afghans facing starvation, Kakar closed a number of his other businesses to focus on the payments platform.

The context that made HesabPay necessary: a paralyzed banking sector, frozen assets, sanctions, and a shortage of physical currency. Only about 6 percent⁹ of Afghans have bank accounts, while roughly 60 percent¹⁰ own a feature or smartphone. 'We wanted to create a system that would allow aid organizations to get money to people in need quickly and efficiently, and we wanted to make it easy for people to send and receive money without having to go through a bank,' Kakar told the Algorand Foundation.

In September 2022, with Algorand Foundation grant funding, HesabPay completed its migration from the Stellar network to Algorand — the switch announced by MIT professor Silvio Micali, Algorand's founder. A year later, Algorand Ventures made a strategic investment in the company. HesabPay uses Algorand as its settlement layer, the component responsible for finalizing and recording transactions. 'Transaction fees on Algorand are very low and consistent. Settlement is extremely fast, and the network is never down, making it the perfect solution for our needs,' Kakar said. Algorand's Pure Proof-of-Stake consensus — the mechanism by which network participants agree on transaction order without proof-of-work's energy competition — finalizes blocks in about 4.5 seconds at near-zero cost, which for a humanitarian platform is the difference between a disbursement that arrives while a family still has food and one that arrives after.

The design is fully custodial: upon signup and basic identity verification (KYC), HesabPay creates an Algorand wallet for each user and holds the private keys itself — simpler for users, but requiring trust in the provider. Accounts are tied to phone numbers, secured by PINs and two-factor authentication, and accessible from a smartphone app, via USSD (the text-menu system on basic phones that works without a data connection), or through QR-code cards. The company says it covers all 400¹¹ districts and 34¹² provinces, with locally managed offices for cash-in and cash-out support, and works with Afghanistan Payments Systems, commercial banks, and mobile network operators. Users can buy airtime, pay electricity bills, and send funds to other users.

The money that moves on-chain is HAFN ('Hesab Afghani'), a stablecoin¹³ — a digital token designed to hold steady value against a currency rather than float like Bitcoin — pegged to the Afghan afghani, alongside a dollar-pegged sibling, HUSD. The local-currency peg matters because recipients take no exchange-rate risk at the point of spending; as CGAP later analyzed, this design shifts the foreign-exchange cost to the implementer instead. HAFN is permissioned: it is not listed on public exchanges, HesabPay mints it against local fiat deposits, and its on-chain configuration gives the platform manager and freeze authority over balances.

On-chain records confirm the token is a real, actively used settlement instrument: the ledger showed 355 HAFN transfers in a single recent sampling window dated August 7, 2026, mostly small-value payments between platform-managed accounts. They also confirm its architecture. HAFN's registered supply is set to the protocol's uint64-max 'no fixed cap' sentinel rather than a real economic total — effectively an uncapped, centrally controlled token — so a percentage computed against that registered maximum is not a meaningful concentration figure; on-chain records show real balances spread across roughly 98,000 distinct holder accounts (the largest non-issuer holder recorded at about 134,000 HAFN), and the token is not listed on Algorand's public DEXs. That is the honest technical picture of 'Algorand-powered': the blockchain is the settlement spine and audit log of a centrally managed stored-value system, not an open retail cash float in which recipients hold meaningful balances.

The Numbers Ladder, and the Independent Check

The adoption story is documented in layers. As of mid-2024, HesabPay reported nearly 4,000¹⁴ daily users, 1,700¹⁵ active merchants, over 3.5 million¹⁶ transactions on the Algorand blockchain, and 170,000¹⁷ electric bills totaling $4 million¹⁸ paid monthly, with more than a dozen organizations using the platform. By the end of 2025, the Foundation was reporting that more than a million Afghans had received humanitarian payments through HesabPay during the year — including cash support from the World Food Programme, UNHCR, and the World Bank — that 30 percent¹⁹ of Afghanistan's electricity bills are paid through the platform, and that cumulative transaction value had reached $10 billion. Those last figures come from the Foundation's own announcements and ecosystem channels; the $10 billion cumulative figure in particular is an aggregate claim the company has not publicly broken down. UNHCR's July figures, cited above, remain the only deployed-scale numbers sourced from an aid agency rather than from the Foundation's own releases.

Independent evidence supports the real-world effects. A London School of Economics study²⁰ that tracked 2,500²¹ women using HesabPay found that direct digital aid meant fewer skipped meals for their families; that local authorities refrained from taxing recipients of digital aid; that 98 percent²² of assistance was spent digitally with local merchants; and that delivery costs fall as the operation scales. In 2024, the World Food Programme named HesabPay one of its 'ten groundbreaking innovations.'

What Mercy Corps Measured in Syria

Mercy Corps²³' involvement goes beyond Council membership: its venture arm, Mercy Corps Ventures²⁴, says it has launched 16²⁵ crypto pilots in more than 10²⁶ countries over four years and is now building a Humanitarian Venture Lab. Its flagship is a stablecoin aid pilot in northeast Syria whose results the team published in October 2025.

From August 2024, Mercy Corps Ventures — with local partner Pioneers Innovation for Sustainability and Impact, and HesabPay as the wallet provider — distributed aid to 100²⁷ smallholder farmers in the Al-Hasakah region, reaching an estimated 650–700²⁸ people. The flow shows the architecture at work: funds left the Mercy Corps treasury in the United States as USDC (a dollar-pegged stablecoin issued by Circle), arrived in the Mercy Corps Syria country office's HesabPay wallet, and were pushed to individual farmer wallets, spendable at participating local vendors. The point was to route around a country cut off from the international financial system for over a decade, where sanctions, currency devaluation, and absent banking infrastructure made conventional aid transfers slow and expensive — delivery through international money transfer agents (the licensed firms that move cash across borders) typically took 28 to 45²⁹ days. Mercy Corps' own measurement:

Source: 30 mercycorpsventures.com

The adoption findings mattered because the cultural default was cash — as project staff put it, 'Syrians like to touch physical money' — so a strong preference for digital, driven by safety from fraud, corruption, and theft, plus speed and ease of use, was itself a result. A participating farmer said: 'Better because it's faster. My rights are protected in my account in a secure manner. No favoritism or loss of notes.'

The pilot's own write-up is candid about friction. Local mobile networks could not deliver SMS verification PINs — the 'biggest surprise' — forcing a workaround of sending codes over WhatsApp. Some older participants needed extra training, and internet gaps were bridged with mobile hotspots in central public areas. Vendors hesitated, and transaction data showed the two participating vendors cashed out almost immediately after large payments, a sign of lingering caution about holding the new money. The environment itself shifted mid-pilot: Syria's regime change in late 2024 and the lifting of sanctions in June 2025 changed the rules the experiment was operating under. A vendor who stayed: 'All the features in the app were very easy to use, clear and without any complications so that both young and old can use it easily. I gained the trust of the people in the region.'

The same model is being tested in Afghanistan. In early 2025, in a program documented by CGAP, the Community Driven Development Organization partnered with HesabPay³² and Mercy Corps Ventures³³ to transfer US$100 per month in HAFN³⁴ to 100 rural households in the remote mountains around Gardez. Because HAFN is denominated in afghanis, recipients faced no foreign-exchange risk at the point of spending; because connectivity is scarce, HesabPay deployed satellite hubs, and households used offline-capable AfPay magstripe and QR cards with static PINs. Mercy Corps Syria is exploring stablecoin aid in other locations for 2026, and ecosystem coverage cites Sudan and Haiti as expansion targets — announced destinations, not yet deployments.

The Council, the Academy, and the Apparatus Around the Mission

The Humanitarian Aid Payments Council is real: an annual convening hosted by the Algorand Foundation, first assembled in 2024, held in Berlin in September 2025, and again in Washington in July 2026, which the Foundation framed as a shift 'from pilot projects to institutional-scale deployments.' Its membership reads like a summit of the movement's ambitions — UNHCR, the World Food Programme, UNDP's Alternative Finance Lab, Mercy Corps Ventures, Mastercard, Visa, Circle, Worldpay, the CALP Network, Coala Pay, Meld, Paycode, Rahat, Quantoz, Polisync, the Disaster Services Corporation, the St. Vincent de Paul Society, COAR Global, Market Impact, and others — with six new members admitted in Berlin (UNHCR, Mastercard, Paycode, Coala Pay, Meld, and Rahat). Government and academic observers attended from the German Federal Foreign Office, the UK's Foreign, Commonwealth and Development Office, and King's College London. In Berlin the Foundation demoed its Aid Trust Portal³⁵ — a compliance tool that screens wallets against blockchain-intelligence databases and flags accounts that have touched suspicious addresses, already in field use with HesabPay and Mercy Corps Ventures — with a live payment from the United States to rural Syria. The Washington meeting was held alongside the Treasurers Roundtable, which gathers NGO treasury leaders responsible for moving billions of dollars in aid annually.

The Council's mandate is coordination and evidence, and its own published takeaways show how far that is from trivial. At the Washington meeting, participants agreed the sector still lacks rigorous, end-to-end data comparing stablecoin rails against traditional systems; that on- and off-ramp infrastructure is the biggest operational bottleneck; that foreign exchange is the most promising area for innovation; and that the sector must move beyond isolated pilot programs toward shared evidence. In other words, after three years, the Council's own conclusion is that the open questions are measurement, liquidity, and FX — not whether the technology can function.

Around the Council sits a larger apparatus. The Foundation's Blockchain Academy for Humanitarian Aid has trained hundreds of UN personnel from more than 50³⁶ countries, run in partnership with UNDP³⁷'s Alternative Finance Lab³⁸, whose first cohort drew interest from 134 countries and which launched a second global cohort in December 2025. A partnership with Paycode³⁹, whose offline-first biometric identity technology reports over 6 million⁴⁰ users across Afghanistan⁴¹, Ghana⁴², Zambia, Mozambique, Sudan, and the Democratic Republic of the Congo, aims to extend blockchain-based services to people with no internet, no formal ID, and no bank account. The Foundation also embedded staff in Syria through 2025, including a dedicated consultant in Damascus.

One contrast belongs in the record: the United Nations is not betting exclusively on Algorand. In July 2026, UNDP announced a separate agreement with the Stellar Development Foundation to make Stellar a standard tool for aid payments across UNDP country offices. Algorand is one of several rails UN agencies are actively evaluating — a fact that does not diminish the UNHCR-Afghanistan numbers but does locate them in a multi-chain market.

What CGAP, the World Bank's Researchers, Concluded

CGAP published 'Stablecoins in Humanitarian Cash Transfers' in June 2026, authored by William Cook, Dylan Lennox, Sara Murray, and Souraya Sbeih. It is the first independent, institutional treatment of exactly the experiments Algorand has been running, and it is notably more sober than the movement's own press.

The report's core finding is that stablecoins restructure costs and risks; they do not remove them. Foreign exchange must happen somewhere in every transfer chain — the question is who bears it. In the via-partner approach, stablecoins settle the cross-border leg while a local disbursement partner converts to fiat and pays recipients as before; in the direct-to-recipient approach, funds reach individuals' wallets and recipients organize their own conversion to cash or goods. The report's comparison:

Across the five criteria it scores — cost, timing, market access, transparency, and last-mile access — recent pilots show the clearest gains in transparency and timing, while recipients in the direct model can face 3–8 percent cash-out costs. The strongest argument for stablecoins is market access: in the report's words, 'In some corridors, the choice is stablecoins or no transfer at all.' The report is chain-agnostic — it describes a UNHCR USDC pilot for Ukrainian recipients on Stellar with MoneyGram as the cash-out network, a CARE program in dollarized Ecuador using Celo's cUSD on NFC cards, and the Mercy Corps/HesabPay HAFN program in Afghanistan as parallel cases — and it lists Algorand's open-source Aid Assist tool among free wallet-provider options for implementers. Its conclusions sit close to the Council's own honest self-assessment, a sign that the sector's internal debate has largely caught up with the researchers. Nothing in the report endorses a chain; the gains it verifies belong to the model, not to any one network.

India: The Financial-Inclusion Wing, at Pilot Scale

Algorand's humanitarian positioning also runs through financial inclusion in India, where two programs target women locked out of formal finance. The first is Mann Deshi⁴³. Its founder, Chetna Gala Sinha, started the microfinance bank for rural Indian women in 1996. The origin story the Foundation tells captures the ethos: when Sinha explained that members needed to create numeric PINs to access their accounts, the women refused and suggested thumbprints instead. 'No one can steal our thumbs,' they said. The blockchain program is cast as a 'digital thumbprint' continuation of that negotiation.

The problem is measurable: over 87 percent of women-owned micro-enterprises in India are excluded from formal loans, per figures Sinha cites, because they lack formal credit histories, collateral, and standard KYC documentation. The solution, launched on Algorand mainnet in 2025, is the Mann Deshi Credit Scorecard, built on what the teams call a Tokenized Trust Framework. Instead of a credit bureau file, the scorecard assembles a portable token-based profile: verified identity; business presence attested by Mann Deshi field staff even for informal enterprises; participation and performance in Mann Deshi's training programs; repayment history with Mann Deshi's internal credit facilities; skills certifications; and 'social collateral' such as savings patterns. Women share tokens that attest to claims — for instance, that average monthly revenue exceeded a stated threshold for a stated period — rather than the full documents, with Mann Deshi acting as the trusted verifier and the underlying paperwork held in DigiLocker, India's government-run document wallet. The design reduces exposure of personally identifiable information rather than adding to it.

The verified scale is modest. The pilot onboarded 500⁴⁴ women, and over half of the first loan applications were approved or disbursed within one month, against several weeks for conventional processing — significant for first-time borrowers, but a pilot. Secondary coverage crediting the program with 400,000 rural women served is not supported by the Foundation's own materials or by any primary source. A parallel effort with SEWA, the Self-Employed Women's Association, has gotten further: its Aadhaar-based (India's biometric national ID) Digital Health Passport has onboarded more than 5,000⁴⁵ informal-sector women workers to blockchain wallets and recorded over 80,000⁴⁶ on-chain transactions, helping them access government health and welfare programs. 'We see our partnership with the Algorand Foundation⁴⁷ as an important step forward in including women in the digital economy, bridging the digital divide,' SEWA director Mirai Chatterjee said at the launch.

Where the Programs Actually Run

Compiling the verified footprint, the distinction between deployed, piloted, and announced is the story:

Source: 48 algorand.co

The impact team's mantra is that it goes 'where money isn't working.' Judged against that standard, the honest ledger is mixed. Afghanistan is genuine institutional deployment with UNHCR as the source of the headline figures and an LSE study as independent corroboration — but it runs on permissioned, custodial rails whose transparency is real at the settlement layer and whose identity mapping lives with the platform operator. Syria is a small, tightly measured pilot whose findings align with CGAP's assessment: timing and transparency gains, costs shifted rather than erased. India is early. The Council and Academy are real institutions, but their progress metrics are self-reported by the Foundation, and the Council's own takeaways concede that rigorous end-to-end evidence does not yet exist. And the sector is multi-chain: UNDP's separate Stellar agreement and CGAP's chain-agnostic case studies both make clear that Algorand is competing for the same aid dollars rather than owning them.

What to Watch Next

Three open questions define the next phase. Whether the Washington summit's commitment to shared evidence produces the end-to-end cost and timing comparisons the Council says are missing — with UNHCR's Afghanistan operation the natural first dataset. Whether the Syria pilot's economics replicate at scale in other corridors, particularly Sudan and Haiti, where the constraints differ. And whether the Indian pilots move past the low hundreds of participants — the gap between a working scorecard and a lending-market standard.

The fair closing note is CGAP's own: for recipients in the most constrained corridors, the practical choice was never between blockchain and banking — it was between stablecoin settlement and no transfer at all. Algorand's Afghan deployment has now demonstrated that claim at a scale of more than 600,000⁴⁹ people, with aid-agency numbers behind it. The next twelve months will show whether the documented efficiency gains survive contact with the hard problems the Council itself has identified — and whether the countries betting on blockchain aid are betting on a settlement rail, or on a story.