Central bank accounts: how settlement actually happens

Settlement in central bank money means that the final transfer of value happens by moving balances between accounts held at the central bank, and understanding those accounts is essential to understanding how any payment system works. A commercial bank holds a settlement account at the central bank, and that account is the place where its payment obligations are finally discharged. When Bank A pays Bank B in a real-time gross settlement system, the central bank debits Bank A settlement account and credits Bank B settlement account, and at that instant the payment is final and irrevocable. No other form of settlement carries the same legal certainty, which is why the PFMI calls settlement in central bank money the safest form of settlement. Why central bank accounts matter is that they are the only accounts that cannot fail. A commercial bank can fail, and a payment settled in commercial bank money, which is a claim on a commercial bank, carries that bank credit risk. A payment settled in central bank money is a claim on the central bank, and the central bank cannot default in its own currency. This is why systemically important payment systems settle in central bank money, and why the European Central Bank T2, the Bank of England CHAPS and the Federal Reserve Fedwire all operate on central bank accounts. The Bank for International Settlements has stated this principle consistently since the 1990s, and it is the foundation of settlement finality law in every major jurisdiction. The parties are the commercial banks that hold settlement accounts, the central bank that operates them, and the payment system that sends settlement instructions. In the euro area, each bank holds a main cash account, called an MCA, and one or more dedicated cash accounts, called DCAs, for specific services such as T2 real-time gross settlement, TIPS instant payments and T2S securities settlement. Liquidity moves between the MCA and the DCAs by camt.050 liquidity transfer messages, so a bank can concentrate its liquidity where it is needed. In the United Kingdom, each CHAPS direct participant holds a settlement account at the Bank of England, and in the United States each Fedwire participant holds a master account at the Federal Reserve. In practice a settlement instruction travels from the payment system to the central bank as a structured message. In T2, the instruction is a pacs.009 financial institution credit transfer, which carries the debtor agent, the creditor agent, the amount, and the settlement date. The central bank validates the instruction, checks that the debtor account has sufficient funds or intraday credit, debits the debtor and credits the creditor, and returns a pacs.002 payment status report confirming settlement. The whole process takes seconds. The central bank also produces a camt.053 statement at the end of the day, so each bank can reconcile its settlement account against its own ledger. A worked example shows the flow. Bank A owes Bank B EUR 25 million in a T2 settlement cycle. The T2 system sends a pacs.009 to the European Central Bank with Bank A as debtor and Bank B as creditor. The ECB checks Bank A main cash account, which holds EUR 40 million, debits EUR 25 million, leaving EUR 15 million, and credits Bank B account, which rises from EUR 10 million to EUR 35 million. The ECB then sends a pacs.002 with status ACCC, meaning settlement completed on the creditor account, to both banks. Bank B can now safely credit its customer, because the funds are in its central bank account and cannot be reversed. What happens when a settlement account has insufficient funds is the key failure mode. The central bank rejects the instruction and returns a pacs.002 with status RJCT and a reason code, and the payment is not settled. The payment system must then decide what to do with the unsettled payment: queue it, return it, or retry it after the bank provides liquidity. In T2, an unsettled payment stays in the queue and is retried when funds arrive; in a deferred net settlement system, the whole cycle may be delayed or the failing bank may be excluded. The operations team must know the exact procedure, because an unsettled payment at the end of the day is a failed settlement, and failed settlements are reportable to the regulator. The operational and accounting implications are that the treasury team must manage the settlement account balance through the day, moving liquidity between the MCA and the DCAs as needed, and the operations team must monitor the account continuously. On the ledger, the settlement account is a balance sheet asset, and every debit and credit must be reconciled to the central bank statement. A difference between the bank ledger and the central bank statement is a break, and breaks in the settlement account are the highest priority reconciliation item in any bank, because they mean the bank does not know its true liquidity position. For a business analyst, central bank accounts are a data and interface subject. You must specify the account structure, the message formats for settlement instructions and status reports, the liquidity transfer mechanism, the cut-off times for each account, and the reconciliation process. You must also specify what happens when an instruction is rejected for insufficient funds, how the system retries, and how the operations team is alerted. A common gap is failing to specify the camt.053 statement reconciliation, which is how the bank proves its settlement account is correct at the end of every day. Takeaway: settlement in central bank money is final because the central bank cannot fail, and every other form of settlement is a claim on someone who can. Further reading: the ECB TARGET services documentation for main cash accounts and dedicated cash accounts; the Bank of England CHAPS reference manual; the Federal Reserve Fedwire services documentation. Last reviewed 7 November 2026. A bank that settles in commercial bank money is trusting the settlement bank, and that trust has a price.