What a Central Counterparty Actually Does

Central counterparties are described as making markets safer. That is roughly true and badly explained. A CCP does not make risk disappear. It gathers risk that was spread thinly across a web of bilateral relationships and concentrates it in one institution built to withstand it.

Whether that is an improvement depends entirely on how well that institution is built — which is why CCPs are among the most heavily supervised entities in finance.

Novation: the mechanism

Two parties agree a trade. Without clearing, they are contractually bound to each other, and each carries the other's credit risk until settlement. In a market with many participants, every firm must assess every possible counterparty.

Clearing replaces this through novation. The original contract is legally extinguished and two new contracts take its place: buyer-to-CCP and CCP-to-seller. The CCP becomes buyer to every seller and seller to every buyer.

The original parties now have no contractual relationship. Neither needs to care whether the other is solvent, because neither is owed anything by the other.

Netting is where the efficiency comes from

Once everyone faces the same counterparty, positions can be netted multilaterally.

A firm that bought a hundred contracts and sold ninety-eight faces the CCP for a net two, not for a hundred and ninety-eight gross obligations. Collateral requirements, settlement volumes and balance-sheet exposure all collapse accordingly.

That compression is most of the commercial case for clearing. It is also why clearing mandates concentrated on standardised products: netting only works across contracts that are genuinely fungible.

How the CCP protects itself

Having absorbed everyone's counterparty risk, the CCP has to survive a member failing. It does so in layers.

Initial margin is collateral posted up front, sized to cover the estimated cost of closing out that member's positions if it defaults. It is a forward-looking estimate of a liquidation cost, not a deposit.

Variation margin is exchanged as prices move, settling gains and losses so that unrealised exposure does not accumulate. This is why a cleared position generates daily cash flows rather than a single settlement at maturity.

The default fund is a mutualised pool that every member contributes to, available when a defaulter's own resources are exhausted.

The CCP's own capital sits in the sequence too, deliberately — so the operator has money at risk before surviving members are touched.

The order in which these are consumed is the default waterfall: the defaulter's margin first, then its default fund contribution, then a tranche of CCP capital, then the contributions of surviving members. It is published, so members can see their exposure to other members' failures before joining.

The international standards governing all of this — margin methodology, default management, governance, recovery planning — are set out in the BIS Committee on Payments and Market Infrastructures' Principles for Financial Market Infrastructures, which treats CCPs as systemically important infrastructure rather than ordinary market participants.

What concentration actually means

The trade is explicit: many small, opaque, bilateral exposures become one large, transparent, heavily defended exposure.

That is usually better. Bilateral webs are hard to see through — during a stress event, nobody knows who is exposed to whom, and the uncertainty itself causes firms to stop dealing. A CCP makes the structure legible and enforces collateral discipline nobody would impose on themselves.

But it creates a single point of failure. A CCP that could not meet its obligations would fail simultaneously against every member of a market. This is why they hold prefunded resources sized against severe scenarios, run default management exercises, and maintain recovery plans — and why their supervision is continuous rather than periodic.

The procyclicality problem

The most substantive criticism is structural rather than a matter of implementation.

Margin models size requirements to observed risk. When volatility rises, required collateral rises. That is correct in isolation and awkward in aggregate: every member is asked for more collateral at the same moment, during precisely the conditions where funding is hardest to obtain.

Firms meet the call by selling assets. Selling into a stressed market moves prices further. The model observes higher volatility and calls for more.

There is no clean fix. Margin that does not respond to risk leaves the CCP undercollateralised when it matters; margin that responds fully amplifies stress. Regulators require anti-procyclicality measures — floors, buffers, longer lookback windows — which dampen the effect rather than remove it. The tension is inherent to collateralising against a moving estimate.

The connection to settlement

A cleared trade still has to settle, and a CCP standing between the parties does not change when the transfer becomes irrevocable. Clearing determines who owes what; settlement moves the value; finality makes it unconditional. Those remain three separate events, as covered in our explainer on settlement finality, and a CCP sits across the first of them.

The question worth asking

Not "is this trade cleared?" but:

• What does the waterfall look like, and where do surviving members sit in it?
• How is initial margin calibrated, and over what lookback?
• What are the prefunded resources sized against?
• What happens after the waterfall is exhausted?

That last one is the least comfortable and most important. Every CCP has an answer, it is written down, and most participants have never read it.

Governance, margin and default-management standards described here follow the BIS Committee on Payments and Market Infrastructures' Principles for Financial Market Infrastructures, read on 1 October 2026 and linked above. Individual CCP rulebooks differ; this describes the common framework, not any specific clearing house.