SWIFT's blockchain ledger and what it means for African remittances
Published · MomoCalc Research Team

Does SWIFT's new blockchain ledger make sending money to Africa cheaper right now? No.
It is a bank-to-bank settlement upgrade being piloted by 17 large banks. It does not touch the mobile-money wallet, the agent, or the withdrawal fee — which is where the cost of an African remittance actually lives. Here is what changed on , and when, if ever, it reaches your transfer.
One quick disambiguation before we start, because two different things are called SWIFT. This page is about SWIFT the network — the co-operative that carries payment messages between banks — and the new blockchain ledger it has just switched on. It is notabout your bank's SWIFT/BIC code, the identifier you need to give someone so they can wire you money from abroad. If that is what brought you here, you want our SWIFT/BIC code lookupinstead. The two are unrelated: the ledger below does not change your bank's code, and your code has nothing to do with blockchain.
What SWIFT actually launched on 9 July 2026
On 9 July 2026 SWIFT said its blockchain-based shared ledger was ready for initial use, with 17 banks across six continents preparing to pilot live cross-border payments using tokenised deposits. Hold onto that phrasing:ready for initial use, and banks preparing to pilot. Reporting on the day noted the first transactions had not yet taken place.
The 17 banks are ANZ, BNP Paribas, BNY, Citi, DBS, First Abu Dhabi Bank, FirstRand, HSBC, Itaú Unibanco, Lloyds, Mashreq, MUFG, OCBC, Standard Chartered, UBS, UOB and Wells Fargo. The ledger is built on Hyperledger Besu, the open-source, Ethereum-compatible framework. Several outlets also report Chainlink's CCIP as the cross-chain interoperability layer, though that detail is not in every account, so treat it as reported rather than settled. SWIFT unveiled the project at Sibos in September 2025 and completed the design phase in March 2026.
The load-bearing caveat. The ledger sits abovethe existing rails; it does not replace them. Tokenised deposits are issued on each bank's own ledger, and the shared ledger records and validates the banks' payment commitments to one another. That lets client funds move around the clock, including overnight and at weekends. But final interbank settlement still runs through the conventional system — RTGS and correspondent banking. Anyone telling you SWIFT has replaced correspondent banking has not read the release.
Is any African bank involved?
One. FirstRand, of South Africa, is the single African bank on the 17-bank roster. There is no Nigerian, Kenyan, Ghanaian, Egyptian or francophone-African bank on the list. And there is no mobile-money operator on it at all — no MTN, no M-Pesa, no Airtel, no Orange, no PSB.
That single fact sizes the whole "what this means for Africa" question down to something honest. One South African bank in a 17-bank consortium is a foothold, not a rollout — worth watching, but not worth telling anyone their remittance is about to get cheaper.
Why this changes nothing at the last mile, yet
Here is the distinction almost every write-up of this launch skipped, and it is the whole story. What SWIFT upgraded is wholesale: how big banks settle with each other. What sets the cost of an African remittance is retail — the last mile. Different layers of the stack, and improving the first does not automatically touch the second.
Make it concrete. A family in Kampala receives money from a relative in London. The money lands in an MTN MoMo or Airtel Money wallet. To turn it into rent or school fees, someone walks to an agent and cashes out — and at that moment they pay the operator's withdrawal fee and, in Uganda, a 0.5% government levy on the cash-out. Before that, the sending service already took its cut, most of it buried in the exchange-rate margin rather than shown as a fee. A settlement ledger shared between 17 large banks does not reach any part of that. It does not make the agent cheaper. It does not remove the levy. It does not narrow the FX margin on the app the sender used. None of those 17 banks is standing between that sender and that wallet.
And the last mile is where the money goes. The World Bank puts the global average cost of sending money at 6.36% of the amount sent (Remittance Prices Worldwide, Q3 2025), with Sub-Saharan Africa the most expensive region in the world to send to, at 8.46% — well above the UN SDG 10.c target of 3% by 2030. That 8.46% is not made of interbank settlement latency. It is made of provider margins, cash-out fees, thin competition on specific corridors, and taxes. A faster overnight settlement rail between global banks leaves every one of those inputs exactly where it found them.
If you want to pay less to send money to Africa today, the levers are unglamorous and they all sit at the retail layer: compare what your recipient actually keeps rather than the advertised fee, on our provider comparison; check the receive-side cost on the corridor you actually use, like UK to Uganda, UK to Nigeria or US to Kenya; and know what the cash-out will cost before your family walks to the agent — for Uganda, that is the 0.5% withdrawal levy.
The problem it doesn't solve: correspondent-banking de-risking
There is a structural reason Africa is the most expensive place in the world to send money to, and it is not settlement speed. Over the past decade, global banks have withdrawn from correspondent relationships they judged too risky or too small to be worth the compliance burden — a pattern regulators call de-risking, and one the BIS and the Financial Stability Board have tracked for years. When the number of banks willing to carry money into a corridor falls, the survivors face less competition, and the price of getting money in goes up.
A faster, always-on settlement layer shared by 17 large banks does not reverse that. If anything it is orthogonal: the banks on this ledger are precisely the ones that never left. De-risking is about who is willing to serve a corridor at all, and at what compliance cost — not how fast they settle once they have decided to.
What would actually matter for Africa
None of this means the ledger is irrelevant forever — it means the thing to watch is specific. The signal that would genuinely matter is the day a major African mobile-money operator, or a large Nigerian or Kenyan bank, connects to this ledger — or the day a payment service bank integrates tokenised deposits into the wallet layer itself, so that the improvement reaches the account a family actually receives into. That is the moment wholesale plumbing starts touching a real remittance. Until then, this is institutional infrastructure: genuinely significant for banks, and genuinely invisible to the person cashing out in Kampala.
We are keeping a standing watch on exactly one thing here — an African-bank or mobile-money connection to the SWIFT ledger — and we will update this page, with the date changed, when it happens. If this page still says what it says today, it means it has not.
Frequently asked questions
Will SWIFT's blockchain make my remittance to Nigeria or Kenya cheaper?
Is this a cryptocurrency or a stablecoin?
When will it reach mobile money?
Which African bank is involved?
- Swift, press release, 9 July 2026 — "Swift's blockchain ledger ready for use as 17 banks set to pioneer tokenised cross-border payments" (the roster, "ready for initial use", six continents).
- Ledger Insights, 9 July 2026 — the settlement mechanics (tokenised deposits on banks' own ledgers; interbank settlement via RTGS or correspondent banking), Hyperledger Besu, and that the first transactions had yet to happen; Sibos September 2025, design phase completed March 2026.
- CoinDesk, 9 July 2026 — banks "preparing to begin testing live transactions"; movement of funds overnight and at weekends "before final settlement through existing payment systems".
- World Bank, Remittance Prices Worldwide, Q3 2025 — global average 6.36%; Sub-Saharan Africa 8.46%, the most expensive receiving region. The same figures we use on our provider comparison.
No African rollout timeline has been announced by anyone, and none is implied here. Where reporting disagreed (for example on the interoperability layer), we say so rather than pick the tidier version.
Last updated . This is an editorial guide, not a fee calculator. Facts are dated because this story will move.