Cat bonds: worth a fresh look

When I first started researching insurance-linked bonds, also known as insurance-linked securities (ILS) or cat bonds, they really looked like the magic asset class, with potential for strong returns uncorrelated with equity markets. Believing in magic is rarely a good starting point for investors, so it is useful to understand the source of this unusually large compensation relative to expected losses.¹

Insurers must prepare for potential losses stemming from extremely rare events. But some natural catastrophe risks – such as US hurricanes and earthquakes – are huge and can’t be fully digested by global insurance markets. ILS allows these ‘peak peril’ risks to be passed on to investors.

The insurer manages a key business risk, while investors can earn additional returns for a risk they have not previously been exposed to.

Aversion to negative skew is a key driver for risk premia across many asset classes that we invest into, and buyers of ILS need to accept less liquid securities with heavily negatively skewed returns.² When building multi-asset portfolios, we are concerned that these negatively skewed returns on different assets will cluster together. But what makes ILS so interesting is that natural disasters only trigger broader financial contagion and link to the global economic cycle in the most extreme circumstances (e.g. the Fukushima disaster in 2011).

ILS goes mainstream

Not many investors knew the asset class when I began researching it in 2009, and that was a key reason for its attractiveness. Over the years, I have been asked why we don’t allocate to these bonds and my response has been that the initial magic has worn off.

While still uncorrelated with equity markets, the expected returns had fallen dramatically as ILS got more popular.

Many institutional investors had picked up on the asset class’s favourable characteristics and driven large inflows, while the capitalisation of the reinsurance industry had improved through a period of low losses.

In 2010, a bond with an ‘expected loss’ (EL) of 2% (as predicted by sophisticated actuarial models) paid a spread of around four times the EL (8% in total, a risk premium of 6%). By 2015, that same risk only attracted a spread of around 2.5 times EL (5% total, a risk premium of 3%). The reward no longer appeared particularly attractive compared with other asset classes given the bonds’ complexity, limited liquidity and model risks.

A new chapter in ILS

But things have changed a lot recently, and the asset class has seen a huge reversal in pricing. Spreads increased gradually as insurance losses picked up in 2017 and weakened the capital base. We then saw a massive jump in the second half of 2022 for various reasons:

1. 2022 loss experience: there was only one noteworthy hurricane, but ‘Ian’ still caused a huge USD60 billion insured loss and further weakened the capital base of the insurance industry.
2. High inflation: Increases in the values of insured items pushed up the notional demand for insurance.
3. Higher yields: In the low-yielding environment of 2009 to 2020, many investors ventured out of safe assets into riskier ones. Spreads compressed everywhere. As risk-free yields have gone up to a level not seen for a decade, I expect the search for yield and spread compression to revert.
4. Portfolio rebalancing: many investors are now overweight alternatives, including ILS, having suffered losses in both equities and bonds during 2022 and having sold liquid assets to de-risk or to raise liquidity.

At current pricing levels, investors are once again being paid well for holding an asset class that should show strong diversification benefits.

In my view, this combination makes ILS once again a worthy addition to a multi-asset portfolio. 

1. See CFA Institute or American Academy of Actuaries for a more detailed asset class summary.

2. Illiquid asset classes include property and private equity. Lots of asset classes, including credit and listed equities, can also have negatively skewed returns. Neither illiquidity nor negative skew is a problem when managed sensibly in a portfolio context