What changed for swaps after 2008?

Before 2008, the swap market was a web of private bilateral promises. Nobody — not the regulators, not the participants — held a map of who owed what to whom. When one large counterparty wobbled, the honest answer to "who is exposed?" was unknown, and unknown is what turns a loss into a panic.

The response was the largest structural reform in the history of derivatives markets.

The four pillars

Following the G20's 2009 commitments, jurisdictions built broadly similar frameworks — the US through the Dodd-Frank Act, the EU through EMIR, with parallel regimes elsewhere. Details differ by jurisdiction; the architecture is common:

• Central clearing. Standardised swaps must be novated to a central counterparty (CCP). The CCP steps into the middle: it becomes the buyer to every seller and the seller to every buyer. Your counterparty is no longer a bank whose health you must assess — it's the clearing house.
• Margin, on a schedule. Cleared trades post initial margin (a buffer sized to a stressed move) and variation margin (the daily mark-to-market, in cash). CCPs also maintain default funds and a defined waterfall for absorbing a member failure.
• Trade reporting. Transactions are reported to trade repositories, so supervisors can, in principle, see aggregate positions rather than guessing.
• Organised execution. Many standardised swaps must be traded on regulated platforms — SEFs in the US, MTFs and OTFs in the EU — instead of by phone.

A fifth pillar arrived later: margin requirements for swaps that are not cleared, phased in internationally through the second half of the 2010s. The design intent is explicit — make bilateral trading carry a collateral cost, so clearing is the cheaper default.

Did it work?

Partly, and measurably. Central clearing went from a minority practice to the norm for interest rate derivatives — industry and BIS data put the cleared share of interest-rate derivative notional at roughly three-quarters or more. Daily variation margin means exposures are settled continuously rather than accumulating quietly for years.

What the reforms did not do is eliminate leverage. Two honest caveats:

• Not everything is cleared. Single-name equity total return swaps — the Archegos instrument — largely remained bilateral and uncleared. The 2021 collapse happened after a decade of reform, inside the part of the market the reform deliberately left bilateral.
• Clearing concentrates risk as well as managing it. A CCP is a single point through which enormous exposure flows. It is heavily regulated and heavily margined for exactly that reason, and CCP resilience is an active supervisory topic rather than a settled question.

Why this matters for the next unit

Central clearing exists because of one lesson learned expensively: when your counterparty can fail, the quality of your counterparty is part of your position. Hold that thought. Unit 3 covers a product where there is no clearing house at all, and the counterparty is the firm that also quotes your price.

In the data

The post-crisis change is visible in public numbers. The table below is the CFTC's weekly count of credit default swaps outstanding, split by whether the trade sits behind a clearing house.

Work out the cleared share of the total. That split is published only for the market as a whole: there is no view by contract or by counterparty, and the figure is gross notional, not money at risk. The shift to clearing shows up as a share of notional and at no finer resolution than that.

Try it now

1. The table below reads the US 10-year Treasury yield on six days of the second half of 2008. Note the size of the whole half-year move, and the size of the biggest single-day moves inside it.

2. Estimate what daily variation margin those moves would have generated on a $100,000,000 ten-year swap using the sensitivity table from Unit 1.
3. Ask the design question: does moving that cash daily make the system safer, or does it just make the demand for cash arrive faster? Both are defensible — hold both.