WhitmanTrading

What Is Central Counterparty Clearing?

Central counterparty clearing is the arrangement in which a single institution becomes the counterparty to both sides of every trade in a market, replacing many bilateral exposures with exposures to one hub. It reduces the total amount of risk in the system through netting and concentrates the remainder in one place.

Before central clearing, everybody in a market owed everybody else. Central counterparty clearing replaces that tangle with a wheel, and understanding why that helps requires looking at the arithmetic of netting.

How it works

A price series with one institution facing both sides.
One institution faces both sides of every trade. Illustrative chart - not real market data.

Every trade is novated. The original contract is legally replaced by two contracts, each with the central counterparty, so the original parties have no relationship with each other.

A steady series where a network becomes a hub.
A web of exposures becomes a hub. Illustrative chart - not real market data.

The topology changes completely. Instead of every participant assessing every other, each one assesses a single institution, and that institution assesses all of them.

A rising series where offsetting positions cancel.
Netting cuts the total amount owed enormously. Illustrative chart - not real market data.

And offsetting positions cancel. A member long in one trade and short in another has one net obligation rather than two gross ones, which shrinks the total outstanding dramatically.

A falling series with margin and a default fund.
Margin and a default fund cover what remains. Illustrative chart - not real market data.

The netting arithmetic is the whole argument

A choppy series where concentrated risk remains at the hub.
The risk is concentrated, not deleted. Illustrative chart - not real market data.

Bilaterally, A owes B, B owes C, C owes A. Each pair carries its full gross amount, and each party must hold capital against each exposure separately.

A slow series where the structure persists across cycles.
And different again over a long horizon. Illustrative chart - not real market data.

Centrally, all three face the hub, their offsetting obligations cancel, and only the net differences remain outstanding.

A calm series where net exposures are small.
A quiet stretch hides what it measures. Illustrative chart - not real market data.

The reduction is not marginal. In large derivatives markets, netting removes the great majority of gross notional, and that is the single largest safety benefit central clearing provides.

A worked example

Three dealers each trade with each other. A owes B 100, B owes C 100, and C owes A 100.

Bilaterally, 300 is outstanding. Each dealer must fund its obligation and hold capital against the counterparty that owes it.

A falling series with a stop level marked.
A stop fills where the market is. Illustrative chart - not real market data.

Centrally, every position faces the hub — and each dealer’s obligations net to zero. Nothing is outstanding at all.

Now let B fail. Bilaterally, C is exposed to B and must still pay A, creating a chain. Centrally, B’s positions are closed out against its own posted margin and nobody else’s balance sheet is touched unless those resources are exhausted.

What the hub does to survive a default

It closes the defaulter’s positions. Usually by auctioning them to surviving members, which is why membership carries an obligation to bid in such auctions.

It applies resources in a published order — the defaulter’s margin, then their default fund contribution, then a tranche of the hub’s own capital, then the surviving members’ contributions.

The hub’s own capital sitting third is a deliberate incentive. It means lax margin setting costs the institution directly, before it costs the members.

And the layers beyond that exist and are unpleasant. Cash calls on members, and in the extreme haircutting the gains owed to winning positions, are the tools available when the waterfall is exhausted. They are published, they have essentially never been used at scale, and pretending they do not exist is the only dishonest way to describe the structure.

The original data

On this site’s shared series the median bar range is 0.493, the ninetieth percentile 1.101 and the largest single bar 2.338 — the largest being 4.7 times the median.

Margin calibration lives inside that ratio. Cover the median and you are undercapitalised nine days in ten; cover the extreme and the market becomes too expensive to use. Every central counterparty picks a point between them and publishes the methodology.

A candlestick chart annotated with the cost of a round trip.
A round trip costs a share of a bar. Illustrative chart - not real market data.

And the drawdown figures show why a single-day calibration is not sufficient: 95% of bars sit below a prior peak, with a longest stretch of 73 bars. Resources have to be held across periods, not just across days.

A price series with volume shown beneath.
Volume and price measure different things. Illustrative chart - not real market data.

The concentration question

A central counterparty is, by construction, too important to fail. If it stops functioning, the market it clears stops functioning, and several such markets underpin the wider financial system.

Which is why they are regulated as critical infrastructure, with capital requirements, stress testing and recovery plans that go well beyond what an ordinary financial firm faces.

The honest summary is that the tangle was replaced with a knot. The total risk is lower, the transparency is far better, and what remains sits in a small number of institutions whose failure would be worse than any single bilateral default would have been.

That is a trade the system made with its eyes open after 2008, and it is the reason systemic risk discussions now spend as much time on clearing houses as on banks.

Why it was mandated rather than chosen

Bilateral markets did not adopt it voluntarily. Netting benefits every participant collectively and costs each one individually — margin has to be posted, default fund contributions made, operational connections built — so nobody moved first.

The 2009 G20 commitments forced the issue for standardised over-the-counter derivatives, and national regulation followed over the next several years.

The mandate’s own logic is worth noting. It did not claim clearing removes risk; it claimed the bilateral web was unmeasurable, and that a measurable concentration is easier to supervise than an opaque tangle.

That is a judgement about knowability rather than about safety, and it is a more defensible claim than the one usually made on its behalf.

When it fails

The characteristic failure is a member who cannot fund margin while perfectly hedged. Their positions offset economically, but the losing leg demands cash today while the winning leg pays later or sits at a different venue. The member is solvent on any measure of net worth and illiquid on the one measure that matters at that moment. Clearing converted credit risk into liquidity risk, which is a better risk to have and is not no risk — and firms have failed inside that gap while being right about everything else.

A candlestick series with a gap through a level.
A gap skips the level entirely. Illustrative chart - not real market data.

A second failure is assuming clearing covers the whole market. Bespoke bilateral contracts remain in volume, and they carry the exposure clearing removes.

A third is treating a central counterparty as incapable of failing. It is very strong and its resources are finite and published.

A fourth is ignoring procyclicality. Margin requirements rise in stress, demanding cash exactly when it is scarcest.

A declining series cut short at a decision point.
Every member called at once. Who funds it? Illustrative chart - not real market data.

And a fifth is forgetting that members are on the hook for each other. A default fund contribution is a contingent liability to losses somebody else caused.

Clearing house covers the institution and its daily mechanics. Counterparty risk covers what the arrangement is built to remove. And systemic risk covers what concentrating it in one place creates.

What I actually do

The interesting question about central clearing is not whether it helps — it plainly does — but what we have chosen in exchange. We replaced many opaque connections with one very visible one, and decided that a known concentration beats an unknown tangle. I think that is right and it is a choice, not a solution.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.