What tail pricing can and cannot promise

Last week I said tail pricing needs a distinct discipline. The follow-up is what that discipline can and cannot promise you today.

Six sourced loss episodes across 1974-2022 are limited direct evidence for any tail price. Extreme value theory gives you a principled way to extrapolate from a thin record, without creating certainty. A parametric or indemnity trigger gives you a mechanism to pay out, while leaving a design question about how closely the payout matches the real loss. Every piece of this is calibration into a headwind.

An imperfect price with its limits stated, tested and improved, is a better foundation for transferring a loss than relying only on mutualisation or other after-the-fact loss allocation. You can't sell a price you never had to defend.

We put the full sourced account behind this in writing rather than leaving it at a LinkedIn-length summary, linked below.

Read the full sourced account: Cat bond trigger mismatch

This first appeared on LinkedIn on 16 September 2026. If you want to comment or discuss, that's the place.