Cat bond course

A cat bond begins with a special purpose vehicle (SPV) — a legal entity created solely to sit between the sponsor (typically an insurer or reinsurer) and the capital markets. The SPV issues notes to investors and uses the proceeds, together with the premium paid by the sponsor, to fund a collateral account. That account is invested in high-quality, short-duration assets, and it is this structure that makes the cat bond fully collateralised — there is no counterparty credit risk in the way there is with traditional reinsurance.

Investors receive a coupon made up of two components: a risk-free rate (typically SOFR) plus a risk spread that compensates them for bearing catastrophe risk. The spread reflects the probability and severity of loss embedded in the bond's structure. If no qualifying event occurs during the bond's risk period, investors receive their full principal back at maturity — usually after a three- or four-year term.

If a trigger event does occur, the collateral is released — partially or fully — to the sponsor. Payouts can be binary (all or nothing) or proportional (scaling between the attachment point and the exhaustion point). The course explains both, and walks through how the calculation agent determines whether a trigger has been breached.

Extension risk is another concept covered in detail. Some bonds allow the maturity date to be pushed back if a qualifying event has occurred but losses are still developing. This is more common with indemnity triggers, where final loss figures can take months or years to settle.

An indemnity trigger pays out based on the sponsor's actual losses. It offers the closest match between the bond payout and the sponsor's economic exposure, which means low basis risk for the sponsor. The downside: it requires full disclosure of the sponsor's book, slows the settlement process, and introduces moral hazard concerns for investors. Despite this, indemnity triggers dominate the market today.

A parametric trigger pays based on a physical measurement — wind speed at a weather station, earthquake magnitude at a given depth, or sea surface temperature. Payouts are fast and transparent, but basis risk can be significant: a hurricane may cause major insured losses while the recorded wind speed at the reference station falls just below the trigger threshold.

An industry loss index trigger pays based on total insured losses reported by an industry body such as PCS or PERILS AG. Basis risk sits somewhere between indemnity and parametric: the payout depends on the market's loss rather than the sponsor's own loss, so a sponsor with an unusual book composition can find itself out of step with the index.

A modelled loss trigger uses a catastrophe model to estimate what the sponsor's losses would be, given the physical parameters of the event. This offers a middle path — faster than indemnity, less basis risk than pure parametric — but introduces model risk.

Understanding these trade-offs is essential for anyone evaluating cat bonds, whether from the sponsor side (choosing which trigger to use) or the investor side (pricing the basis risk embedded in each type).

Expected loss (EL) is the foundation. It represents the average annual loss the bond is expected to experience, expressed as a percentage of notional. EL is derived from the cat model output and sits between the attachment point (where losses begin eroding principal) and the exhaustion point (where the entire principal is lost).

The spread investors receive above the risk-free rate can be decomposed into expected loss, expense load, and risk load. The risk load is the actual compensation for bearing catastrophe risk. Two standard metrics measure the relationship between spread and risk: the net risk multiple (NRM) — how many multiples of expected loss the investor earns in spread — and the multiple of expected loss (MEL), which includes the full coupon.

The course also covers the Wang Transform, a probability distortion method used to move from physical (objective) loss probabilities to risk-adjusted (market) pricing. This is one of the more technical sections, but it connects the cat model output directly to the price investors see in the offering circular.

Seasonality affects pricing: bonds covering North Atlantic hurricane risk trade at wider spreads during wind season and tighter during the off-season. The secondary market reflects this, and the course explains how bonds are valued on a mark-to-market or mark-to-model basis, depending on liquidity and the proximity of a loss event.

Insurers and reinsurers issue cat bonds for several practical reasons. A cat bond provides fully collateralised, multi-year coverage — eliminating the renewal risk that comes with annual reinsurance contracts. For European sponsors operating under Solvency II, cat bonds can deliver solvency capital requirement (SCR) relief because the collateral structure removes counterparty default risk from the calculation. Repeat issuers such as large global reinsurers use cat bonds as a structural part of their retrocession programmes.

The course also covers sovereign cat bonds — instruments issued by governments and multilateral organisations to pre-fund disaster response. The World Bank (IBRD) has acted as arranger for bonds covering earthquake risk in Mexico, hurricane risk in the Caribbean through CCRIF, and drought risk in Africa through ARC. These bonds typically use parametric triggers, where speed of payout matters more than precision of loss estimation.

The 2017 HIM season (Hurricanes Harvey, Irma, Maria) is used as a case study. Several cat bonds were triggered, giving learners a concrete view of how attachment points, trigger types, and extension periods interact during a real loss event. The course examines which bonds paid out, which extended, and what the recovery experience looked like for investors.