CLS and the ghost of Herstatt
Posted: Wed Oct 28, 2026 6:18 am
In 1974, a German bank called Herstatt was shut down after receiving Deutsche marks from counterparties but before paying out dollars in New York, because of time zone differences. Counterparties lost money despite having done their part of the deal. The episode gave its name to settlement risk in FX: the danger that one side of a trade pays while the other doesn't.
The industry's response, decades later, was CLS Bank, which settles both legs of a trade simultaneously, a mechanism called payment versus payment. Most large trades in major currencies now settle through CLS, dramatically reducing the risk.
Why bother with this as a scalper? Because it explains parts of market structure that otherwise look arbitrary: why settlement takes two days for most pairs, why some currencies settle differently, and why rollover and value dates matter.
It also reminds that FX has real-world plumbing that can fail. Exotic currencies outside CLS carry settlement risk, which partially explains their wider spreads.
Retail traders don't face settlement risk directly, but they face broker counterparty risk, which is a cousin of the same idea.
Is counterparty risk part of broker selection, and has the history of settlement failures changed the view?
The industry's response, decades later, was CLS Bank, which settles both legs of a trade simultaneously, a mechanism called payment versus payment. Most large trades in major currencies now settle through CLS, dramatically reducing the risk.
Why bother with this as a scalper? Because it explains parts of market structure that otherwise look arbitrary: why settlement takes two days for most pairs, why some currencies settle differently, and why rollover and value dates matter.
It also reminds that FX has real-world plumbing that can fail. Exotic currencies outside CLS carry settlement risk, which partially explains their wider spreads.
Retail traders don't face settlement risk directly, but they face broker counterparty risk, which is a cousin of the same idea.
Is counterparty risk part of broker selection, and has the history of settlement failures changed the view?