Cat bond returns per-unit of risk diverge as industry-index deals slide lower: Man Group

Over time, the returns per-unit of risk from catastrophe bonds have, on average, declined faster for industry-index deals, resulting in a divergence between the at-issuance spreads delivered by cat bonds featuring industry-loss triggers versus the rest of the market, data from Man Group shows.

This is a trend that has been evident in the catastrophe bond market for a number of years now, with the multiples-at-market paid at issuance for industry-loss index trigger cat bonds tending to come in lower than other types of deals, particularly indemnity trigger cat bonds.

As we’d written before in 2024, spreads in the outstanding cat bond market declined (or tightened) much faster for industry-loss index trigger catastrophe bond instruments.

New data, shared with us by global independent alternative and active investment management firm Man Group, shows that the at-issue pricing of catastrophe bonds has been diverging for a number of years, with industry-index trigger deals settling at softer pricing.

Man Group’s data shows the cat bond return per-unit of risk over time, based on exponentially weighted moving averages since May 2006, with certain outlier transactions removed from the data set for clarity.

Man Group has tracked the rolling mean at-issue spread multiples for all cat bond deals, versus just industry-index cat bonds, and for all cat bonds excluding the industry-loss trigger component.

Speaking with Artemis, Andre Rzym, partner and portfolio manager at Man AHL, the firm’s systematic and quantitative trading unit, explained the rationale behind the data set and what it reveals.

First the data, for you to review, with the commentary to follow.

This chart displays the ratio of cat bond spread at issuance to expected loss (the multiple), while the solid lines display the rolling averages for each category of cat bonds. It illustrates that industry index cat bond deals have priced increasingly tightly over the last five years, a phenomenon which Man Group noted is also evident in the secondary market.

Rzym told us that using the at-issue data helps to control for impaired cat bonds, which might skew any rolling average, while this data set also excludes certain outlier deals, such as working layer transactions or those where the expected loss is so low it skews the multiple.

Rzym provided some additional commentary to accompany this chart, explaining, “For these reasons, we examine how the average deal spread per unit of risk (i.e. deal spread divided by expected loss) at the point of issuance has evolved over time.

“Excluding outlier deals (i.e. working layer deals, or where the expected losses are so low that pricing is driven by differing costs of capital), the results are plotted in the attached. The dots represent individual new deal issuances, and the solid lines represent the rolling averages (industry index only, all deals, all deals ex industry index).

“The first few years are somewhat noisy due to sparse data, but from around 2010 to 2020, there is not much to choose between the different triggers. From around 2020 onwards, we see industry index deals being launched at progressively tighter risk-reward levels relative to other trigger types.

“Now it may be that there is, a priori, a reason to prefer industry index deals over indemnity deals (e.g. relating to lack of idiosyncratic dependence upon sponsor claims processes) however the steady widening of the gap that we observe is more likely to be driven by supply and demand. For example, some funds may be constrained by mandate to hold industry index deals. Also, given that industry loss deals account for around 20% of the market (as at July 2025), larger funds may also find it challenging to avoid them, even if they wanted to.”

The data, in the chart further up this article, clearly shows that this divergence in return per-unit of risk between industry loss index trigger cat bonds and cat bonds using other trigger types began in 2020.

While a slight divergence had been visible through the softening market years of 2013 through 2018, the gap had narrowed around the bottoming out of that soft market, but began widening again and more meaningfully after 2020.

That widening of the return per-unit of risk gap in cat bonds became more prevalent over the last three years, seemingly being at its widest at the time insurance risk spreads of catastrophe bond issues were at their highest around 2024.

Since when the gap has proven relatively stable and has persisted through the price moderation seen in 2025 to-date.

As we’ve explained before in discussions about industry-loss cat bonds, these instruments have tended to move in and out of favour with some ILS fund managers and investors.

When they have priced more tightly, comparable to indemnity cat bonds, some managers have opted to underweight them, feeling the correlation possible (due to similar index triggers often being used) can make them relatively less attractive at those times.

Conversely though, there are some insurance-linked securities (ILS) fund managers that find industry-index cat bonds relatively more attractive, believing them to be more transparent for modelling purposes, given the availability of data and models to approximate industry loss impacts, as well as the fact they remove any uncertainty around sponsor claims processes that can be a feature of indemnity trigger structures.

There are also a number of cat bond strategies that have a focus on index-trigger arrangement investments, which can result in dedicated capital being made available to support sponsors that opt for this type of structure, which can also serve to drive keener pricing at issuance, depending on market dynamics at the time.

Man Group’s data set provides a helpful way to visualise this trend and the clear divergence that has occurred between spreads of cat bonds with different trigger types since 2020.