How CBDCs Work

The Two-Tier Model

Almost every central bank exploring a CBDC has settled on a two-tier model — and the Bank of England is no exception. In the two-tier system:

• Tier 1 (Central Bank → Intermediaries): The Bank of England issues digital pounds to regulated intermediaries — banks, fintechs, payment processors
• Tier 2 (Intermediaries → Public): Those intermediaries provide wallets and payment services to households and businesses

This means you wouldn't have an account at the Bank of England. You'd have a wallet at your bank or a fintech, but the money in it would be a direct liability of the BOE — not of the intermediary. This is the key difference from a normal bank deposit.

Why Two Tiers?

A single-tier model — where the central bank onboarded every citizen directly — would be a logistical nightmare. The BOE would need to run tens of millions of retail accounts, handle KYC (know-your-customer) checks, deal with lost passwords, fraud and customer service. That's not what central banks are built for. The two-tier model lets the BOE focus on the core infrastructure while private-sector intermediaries handle the customer-facing side — which is what they're good at.

Accounts vs Tokens

There are two fundamental ways to design a CBDC's underlying representation of money:

Account-Based

Each user has an account with a balance, and payments transfer value between accounts. This is how your bank works today. The system knows who owns what, and every transaction is tied to an identity. Pros: straightforward, easy to regulate, easy to trace. Cons: requires always-on connectivity, and every transaction is visible to the central system.

Token-Based

Money is represented as digital "tokens" — like digital banknotes. Ownership is determined by who holds the token (using cryptographic keys), not by a central account register. Tokens can be transferred peer-to-peer, potentially offline. Pros: closer to physical cash, supports offline payments, can offer stronger privacy. Cons: if you lose your keys, you lose your money (like losing a banknote); harder to trace.

The BOE's Approach

The Bank of England's design favours a hybrid approach. The core ledger would track balances, but the system would also support offline payments using a token-like mechanism. This would let two phones exchange digital pounds without an internet connection — similar to handing over a physical banknote — with the transaction syncing to the central ledger when connectivity is restored.

Distributed Ledger Technology (DLT) vs Conventional Infrastructure

Many people assume a CBDC must use blockchain. In practice, most central banks — including the BOE — are not committed to DLT. The BOE's design phase is evaluating both:

• Conventional centralised database — the BOE runs a single ledger. Faster, simpler, and well understood. This is the likely choice.
• Permissioned DLT — a blockchain where only authorised participants (the BOE and intermediaries) can validate transactions. More resilient and transparent, but slower and more complex.

The BOE has indicated that a conventional database is the probable foundation, with some DLT-like features (cryptographic signatures, audit trails) where they add value.

Offline Payments

Offline payment capability is one of the hardest problems in CBDC design. The goal: two people should be able to exchange digital pounds using their phones, with no internet connection on either device, without the risk of double-spending.

Proposed solutions include:

• Hardware-secured elements — phone chips that securely store a limited amount of digital cash and enforce spend limits
• Deferred settlement — the transaction is recorded locally and synced to the central ledger when connectivity returns
• Pre-paid sub-wallets — users load a small offline balance from their main wallet, which can be spent without connectivity

Programmability

One of the most interesting features of a CBDC is programmable money — the ability to attach conditions to payments. For example:

• Government benefits that can only be spent on essentials
• Automatic tax collection at the point of sale
• Smart contracts for escrow or conditional payments
• Time-limited money (e.g. stimulus payments that expire after 6 months)

The BOE has been cautious about programmability, emphasising that the digital pound itself would be a general-purpose payment instrument. Programmable applications could be built on top, but the core digital pound would not have built-in spending restrictions.

Interoperability

The digital pound would need to interoperate with existing payment systems:

• Faster Payments — the UK's real-time bank transfer system
• CHAPS — high-value same-day payments
• Bacs — direct debits and standing orders
• Card networks — Visa and Mastercard
• Cross-border CBDCs — potential future links to the digital euro or other CBDCs

The Bottom Line

CBDCs are complex financial infrastructure projects, not crypto experiments. The UK's digital pound would use a two-tier model with regulated intermediaries, likely on a conventional database rather than a blockchain. The hardest problems — offline payments, privacy, and preventing bank runs — are still being worked out. But the architecture is well understood, and central banks worldwide are converging on similar designs.

Next Steps

• Digital Pound Design — the BOE's specific platform model and holding limits
• Privacy and the Digital Pound — how the BOE is approaching the privacy question
• Digital Pound vs Cryptocurrency — why CBDCs are fundamentally different from crypto

Nothing on this site is financial advice. All content is for educational purposes only. Back to all guides