Clearing Houses and Central Counterparties

A clearing house is more than a calculator. It sets the rules for participation, defines file formats and message standards, enforces cut off times, manages exceptions and provides the audit trail that lets every participant reconcile its own records against the central record. In retail payment systems, the clearing house is often owned by the participating banks or by the central bank, and it operates as a utility that serves the whole market. Its job is to be neutral, reliable and transparent, because every participant depends on its calculation being correct. A central counterparty, often called a CCP, goes one step further. It steps into the middle of every transaction and becomes the buyer to every seller and the seller to every buyer. Instead of Bank A owing Bank B directly, Bank A owes the CCP and the CCP owes Bank B. This arrangement is common in securities and derivatives markets, and it is appearing in some payment contexts too. The benefit is that participants no longer face credit risk on each other. They face the CCP, which is heavily capitalised and tightly regulated. The trade off is that the CCP concentrates risk in one place, so its own risk management must be extremely robust. For a business analyst, the distinction matters when designing system requirements. A clearing house model requires strong reconciliation, net position calculation and settlement instruction generation. A CCP model additionally requires margin management, default funds, position monitoring and default procedures. Both models need clear rules for what happens when a participant fails to pay its net debit. That failure procedure is one of the most important parts of the design, because how the system handles a default determines whether a single failure spreads to other participants or is contained. Make it concrete with numbers. A retail clearing house runs one cycle a day for 12 banks. In that cycle the banks submit payments with a gross value of EUR 4.8 billion, and after netting the settlement amounts add up to EUR 320 million. That is the liquidity saving in one number. EUR 4.48 billion that never has to move. The clearing house also enforces a 2 in the afternoon cut off, rejects files that arrive late, and publishes the net positions by 3 in the afternoon so every bank can fund its debit before the 4 in the afternoon settlement. Now take a central counterparty in a securities market. Two members trade, and the CCP steps in as buyer to one and seller to the other, so neither faces the other. Each member posts EUR 5 million of initial margin, and when prices move against one member by EUR 800,000 in a day, the CCP calls EUR 800,000 of variation margin from that member before the next morning. If the member fails to pay a EUR 12 million net debit, the CCP pays the other members from the margin and the default fund, and the market keeps trading. The two models need different requirements, and confusing them is a common design error. A clearing house design needs strong reconciliation, net position calculation, settlement instruction generation, and cut off enforcement. A CCP design needs all of that plus margin calculation, daily margin calls, position monitoring, a default fund, and a default procedure with a named decision maker. When a business analyst writes requirements for either model, the default procedure deserves the most attention. Who decides that a participant has defaulted, what happens to its queued payments, in what order the margin and the default fund are used, and how the other participants are told. A system that calculates nets perfectly but has no tested default procedure is not finished.