Banks blocking crypto payments: what the UK inquiry proves about who sits in the middle of your money

What closed on August 31, 2026, and why it exists

On July 21, 2026, the UK's Crypto and Digital Assets All-Party Parliamentary Group — a cross-party group of parliamentarians, co-chaired by Lord Vaizey of Didcot and Labour MP Gurinder Singh Josan CBE — launched a formal parliamentary inquiry into access to banking services for the UK crypto and digital-asset sector. The call for written evidence ran six weeks and closed on August 31, 2026. The APPG will publish findings and recommendations to Government.

The scope, as set out by the APPG, covers two things that are often conflated: access to bank accounts and banking services for these businesses, and the restrictions banks place on crypto-related transactions. It asks how significant the problem is, what it costs in innovation and competitiveness, what drives it — legal, regulatory, commercial or operational — and what solutions exist, including international examples. The framing is deliberately open: the APPG said it believed it was "the right time to examine whether legitimate firms continue to face unnecessary barriers to the banking services they need to operate and grow."

This did not come out of nowhere. On August 11, 2026, the two co-chairs wrote to the chief executives of every major UK bank asking them to explain their approach to serving crypto and digital-asset businesses, noting that firms struggle to open accounts and that several banks have introduced restrictions on crypto-related payments. In March 2026, the Economic Secretary to the Treasury told Parliament that the Government would not expect authorised crypto firms to face banking restrictions simply because of the sector they belong to.

So: a documented problem, a ministerial statement that it shouldn't be happening, and a formal inquiry that has now closed its evidence window. What did the evidence actually say?

The number underneath: 40%, £1 billion, and zero explanations

The figure that drove the political attention comes from a January 2026 industry report, "Locked Out: Debanking the UK's Digital Asset Economy," produced by the UK Cryptoasset Business Council from responses by ten of the country's largest centralised exchanges. Reporting on the survey sets out the headline findings:

• ~40% of transactions destined for digital-asset exchanges are blocked or delayed by major UK retail and commercial banks.
• One exchange alone reported close to £1 billion (about US$1.2 billion) in declined transactions over a year, across card payments and bank transfers visible to the platform, in the UK alone.
• Every exchange surveyed said the bank offered no explanation when a payment was declined.
• 80% reported more customers hitting transfer blocks during 2025; around 70% described the banking environment as increasingly hostile, and said it discouraged investment, hiring and product launches in the UK.
• Restrictions are frequently applied without distinguishing an FCA-registered UK business from a higher-risk offshore platform.

The Council's legal argument is worth noting because it is narrow and testable. It points to regulation 105 of the Payment Services Regulations 2017 — "Access to bank accounts" — which requires credit institutions to grant payment service providers access to payment account services on an objective, non-discriminatory and proportionate basis, extensive enough for them to provide payment services "in an unhindered and efficient manner," with transparent criteria and a duty to notify the FCA of refusals or withdrawals. The complaint, in other words, is not that banks assess risk. It is that a categorical rule applied to a whole sector is not an assessment.

The direction of the block is the whole story

Here is where most coverage stops and where a business owner should keep reading, because the mechanics matter more than the headline.

Almost every number above describes money moving in one specific direction: out of a retail customer's bank account, toward a crypto platform. That is the on-ramp — a consumer buying tokens, or funding an exchange account. It is the leg where a bank is unavoidably in the path, because the money starts inside the banking system.

A merchant accepting a stablecoin payment is standing on a different leg entirely. The customer already holds the token. The payment moves from their wallet to yours. No bank authorises it, no acquirer routes it, and no risk team can decline it at 11pm without telling anyone why — because none of them are in the path.

Reading down that table gives the honest shape of the thing. The UK inquiry indicts rows one and four. Row three is the row a merchant actually operates on when they take a stablecoin payment. And row two — the one nearly every business already lives with — has the same structure as row one, but nobody calls it debanking because it has a friendlier name: underwriting.

What this does not prove

It would be easy, and wrong, to turn this story into "banks are unreliable, so take crypto instead." Three limits deserve stating plainly before anything else.

Accepting stablecoins does not cure debanking. A business needs a bank account for rent, payroll, suppliers, VAT and corporation tax. If your bank closes your account, changing how customers pay you does not solve that. What it does is stop the bank's risk appetite from also controlling whether revenue can reach you at all — which is a meaningful separation of concerns, not a cure.

The banks are not acting irrationally. They carry anti-money-laundering obligations, fraud reimbursement exposure on authorised push-payment scams, and supervisory scrutiny. A blanket rule is cheap to run and easy to defend internally. The Council's argument is not that risk management is illegitimate; it is that regulation 105 requires proportionality and case-by-case treatment, and a sector-wide rule delivers neither.

And this is a UK story with a UK legal frame. The pattern is not unique — we have written about Brazil's 24-hour hold on large transfers to self-custody and about what MiCA changed for merchants in Europe — but regulation 105, the FCA and the APPG are British instruments. Don't import the conclusions into another jurisdiction without checking.

What it does prove: three properties worth naming

Strip the politics out and the survey documents three properties of intermediated payments that most business owners have felt but never seen quantified.

1. Discretion without disclosure

The most striking finding is not the 40%. It is that 100% of respondents said no explanation was given. A payment failed, money did not move, and the reason lived inside a system neither party could see. That is the same experience as a card declining at your counter for reasons the terminal will never tell you, or an acquirer holding a settlement batch pending review. The instruction it generates is simple: know which parties can stop your money, and reduce that list where you can.

2. Category treatment beats individual treatment

Restrictions applied without distinguishing an FCA-registered UK business from an unregulated offshore one is the definition of category risk. Any business that has been classified "high-risk" by a payment processor knows the feeling — the classification arrives before anyone has looked at your chargeback rate. It is why we wrote about what a high-risk merchant account alternative actually has to fix: the problem was never your numbers, it was your MCC.

3. The cost lands on the party with no standing to appeal

£1 billion in declined transactions at a single platform is not the platform's loss alone — it is customers who could not do what they intended, and a business that could not book the revenue. When the block is silent, the merchant is left explaining a failure they did not cause and cannot investigate. If your provider disappears entirely rather than declining, the same asymmetry gets worse: we covered that scenario in what happens when a crypto payment provider shuts down.

How Payzum fits the honest version of this

The design conclusion is not "avoid banks." It is keep the number of parties who can stop your money as small as the transaction allows, and be deliberate about the leg where you cannot get that number to zero. That is close to a product spec, and it is the one Payzum was built to.

• Non-custodial, so the processor is not a party that can freeze anything. Funds settle directly to wallets the merchant controls. Payzum never holds, pools or controls the money — the settlement is the payment. There is no Payzum balance, no rolling reserve, no payout schedule to be paused. This is the whole argument for a non-custodial crypto payment processor, and this week's news is the clearest illustration of it we've seen.
• Crypto-only, with optional auto-conversion to USDC or USDT. Payzum accepts crypto and settles in crypto. It does not settle to fiat bank accounts, and it will not pretend otherwise. If you want dollar-denominated value rather than whatever the customer paid with, auto-convert handles it on arrival — you decide once what you hold instead of accumulating a shelf of tokens.
• No chargebacks, because there is no scheme to reverse anything. On-chain settlement is final. That removes the ~120-day reversal window that makes card revenue a receivable rather than money — the mechanics we set out in reducing chargebacks in an online store. The other edge of finality is real too: refunds are payments you initiate, so your policy needs to be written down and honoured.
• Every way of getting paid, without an acquirer. Payment links and buttons with no code; hosted checkout as redirect, modal or inline; invoices with expiry and overpayment detection; recurring subscriptions; a donations or tip page; and POS with a fresh QR per sale, PIN-protected cashiers and per-cashier analytics, where any phone is a terminal.
• Money out on the same rail. Mass payouts by CSV and EVM stablecoin payouts across Polygon, Arbitrum, Optimism, Base, BNB Chain and Avalanche — so paying contractors, affiliates or winners doesn't fall straight back into the leg you just routed around. See crypto mass payouts.
• Chains and speed. Bitcoin, Ethereum, Solana, Polygon, Base, Arbitrum, Optimism, BNB Chain and Avalanche, with typical confirmations around 0.4s on Solana and roughly 2s on Base and Polygon. Plus 2FA, encrypted secrets, signed webhooks and a full audit log, with KYC in the product.

What Payzum does not do, and will not claim: it is not a bank, not a custodian, does not settle to fiat bank accounts, and does not change a single regulatory obligation you already carry.

How it works, step by step

1. Map the gates in your current flow. Write down every party that can stop a payment between your customer's intent and your usable cash: issuer, scheme, acquirer, processor risk team, your bank. Most businesses have never counted them, and the count is usually four or five.
2. Connect a wallet you control and set the settlement policy. Choose which chains and tokens you accept, and whether incoming payments auto-convert to USDC or USDT. This is the one decision that determines what you are holding — made deliberately, in a dashboard.
3. Turn on the instruments that match how you sell. A fresh QR per sale at the counter; payment links for invoices sent by message; hosted checkout on the site; subscriptions for recurring plans; invoices with expiry for B2B terms.
4. Wire it into what you already run. REST API with API keys, signed webhooks and an integration playground, plus drop-in compatibility with existing e-commerce plugins. If you sell an API that AI agents call, x402 lets them pay in USDC on Base per call, straight to your wallet — you configure your existing endpoint and a price; there is no protocol to implement.
5. Be deliberate about the off-ramp. This is the leg that stays banked, so treat it as a designed process rather than an afterthought: know which venue you convert through, keep the relationship documented, and don't let an unplanned conversion be the first time your bank hears about this part of your business.

Where this shows up in real businesses

The abstract argument gets concrete quickly. In each of these, nobody is trying to make a political point about banking — they are trying to stop losing a specific sale.

• The customer whose bank silently declined. An online seller loses a basket because the buyer's card is declined with no reason, twice. The buyer already holds USDC. A payment link settles in seconds to the seller's wallet, and the sale that was going to be written off as "cart abandonment" closes instead.
• The business classified before it was assessed. An operator in a category their processor treats as high-risk carries a rolling reserve and a settlement delay that has nothing to do with their own dispute rate. A non-custodial rail run alongside cards removes the reserve from the portion of revenue that runs on it, because there is no balance held anywhere to reserve against.
• The counter where the payer has no local account. A shop, gallery or charter base sells to visitors whose cards are foreign and whose local bank access is nil. A QR per sale settles to the owner's wallet with no reversal window — the argument in accepting crypto in person and in getting paid from abroad without a bank account.
• The payout that arrives on a bank's schedule. A business paying contractors or affiliates across borders watches transfers sit over a weekend or bounce for compliance review. A CSV batch of stablecoin payouts leaves on the schedule you set, not the schedule your correspondent chain allows.
• The API selling to machines. An AI agent has no bank account, no card, and no capacity to pass a bank's risk model. There is no version of this story where the on-ramp gets fixed for a buyer with no legal personality. That demand is being served today, in USDC on Base.

Intermediated acceptance vs non-custodial acceptance

Objections worth taking seriously

"If banks are blocking crypto, doesn't that make crypto payments harder for my customers too?"

For a customer who has to buy tokens first, yes — that customer meets the same on-ramp friction the survey documents. But the customers who pay merchants in stablecoins overwhelmingly already hold them: people paid in stablecoins, businesses running treasury in them, travellers, and payers in corridors where dollar tokens are already the working currency. Your acceptance rail does not need to solve the on-ramp for the population that already crossed it. It is also worth noting the irony in card-wrapped stablecoin products, which put an intermediary back in the middle — the pattern we examined in "invisible" stablecoin payments.

"Doesn't this just move my risk from a bank to a token issuer?"

Partly, and it should be said rather than glossed over. Holding an issuer's stablecoin is exposure to that issuer's ability to redeem at par, for as long as you hold it. The difference is in the shape: a bank block is a discretionary decision applied to you by category, with no explanation and no appeal; issuer risk is a measurable, disclosed exposure you can size, diversify and shorten by converting on arrival rather than accumulating. We go through it properly in counterparty risk in stablecoin payments.

"Should I drop cards?"

No, and any provider telling you to is overselling. Cards serve the customers they serve well, and a card bundles a transfer instrument with a consumer credit product that stablecoins do not replace. The realistic pattern is both: cards for the mainstream flow, and a non-custodial rail for the sales that cards decline, delay, charge most heavily, or reverse months later.

"Will the inquiry fix this anyway?"

It may improve it. A ministerial statement already exists, regulation 105 already requires proportionality, and the APPG will publish recommendations. But recommendations are not rules, and the timeline runs into the FCA's new authorisation regime rather than ahead of it. Waiting is a strategy with a cost measured in the sales you don't close in the meantime.