Do Recent Events Change the Outlook for Insurance-Linked Security Strategies?

Christine Farquhar

Answers to our clients’ questions about market action and the market environment in a few paragraphs every two weeks.

The strong Atlantic hurricane season and Mexican earthquakes have resulted in a tragic loss of life and property. Insurance-linked securities (ILS), meanwhile, have suffered a less dramatic impact. Estimated insurance losses for 2017 from earthquakes, hurricanes, and other events are between $90 billion and $140 billion. These losses will consume most of the current year’s retained earnings for the whole insurance industry. As claims are settled and estimates refined, the impact on reinsurance contracts will become clear. So far, the effect on yields has been concentrated in US wind, with signs of spreading to the wider market. If the industry does need to replace capital, we expect to see higher headline yields after several years of decline. However, increases are expected to be modest and relatively short lived. This improves the opportunity for long-term investors but, in an illiquid asset class, is scarcely enough to reliably deliver short-term tactical gains.

Catastrophe (CAT) bonds, the most liquid version of ILS, currently offer moderate yields of 4.5%, in line with similar quality (BB/B rated) US corporate bonds. Less liquid private instruments, namely collateralized reinsurance (CRI), offer higher yields between 6% and 10% with less homogenous risks. Their biggest attraction is their low historical correlation with other markets, but limited liquidity, short history, and relative complexity mean they are not for everyone.

Based on official industry estimates, accumulating losses (Hurricane Maria being the latest and potentially most significant) are beginning to put pressure on capital reserves and increasing expectations of higher yields. CAT bonds, which cover more remote risks and pay out only on bigger industry losses, would see modest yield increases. CRI investments generally trade at a premium, pay out on lower aggregate losses, and are more likely to see material increases in yield.

Much depends on rates-on-line (ROL) in the reinsurance markets, which are expected to be higher (up to 125% of last year’s rate) for contracts covering the hardest hit regions of the Caribbean and Southeast United States. ROL on some retrocession contracts, which provide reinsurance for reinsurers (as opposed to the primary insurance companies), could touch 150% of last year. Retrocession is more exposed to “third event” losses like Hurricane Maria and more widespread in private CRI transactions. Wholesale increases in CAT bond yields would depend on cumulative losses exceeding $120 billion. Even so, yields are unlikely to return to historical highs, as the strain on industry reserves is significantly less than in 2011 (the Japan and New Zealand earthquakes) or 2005 (Hurricane Katrina). Instead, yields are expected to reach levels last seen in 2013.

Although CAT bond structures are very similar to corporate bonds, private CRI structures are less standardized. The structure is more akin to the underlying reinsurance market, a key aspect of which is the risk of “collateral trapping.” Uncertainties over final claim settlements allow CRI counterparties to hold on to collateral based on preliminary estimates. Some CRI funds currently have 70% or more of their portfolio trapped in these arrangements, and many ILS strategies are in the 10%–25% range. Even if losses turn out to be far lower than estimated, trapped collateral may miss out on redeployment in the January renewal season.

Investors need to understand how managers treat trapped collateral and calculate fund NAVs (net asset values). If managers side-pocket affected contracts, this reduces liquidity but increases the reliability of the NAV. As losses are finalized, capital is released to existing fund holders. New capital is protected against dilution of returns from trapped collateral unavailable for the annual reinsurance renewal. Alternatively, if managers strike a preliminary NAV on current loss estimates, then investors that trade on this basis might benefit or lose out from future revisions. Insurance claims can take months or years to settle, and the trapped collateral will be a drag on all investors’ returns, so some managers offset this with financial leverage. Quality and transparency of NAV calculations, plus long-term liquidity and leverage of the fund, are critical factors differentiating ILS managers.

Recent events have provided a meaningful stress test for ILS. To date, most managers have performed as expected. Based on estimated insurance industry losses, ILS returns for 2017 will likely range from flat to down 20%, depending on contract exposures and upfront yields. Statistically, 2017 has not been such a rarity as it may have seemed from the headlines. Estimated losses are, nevertheless, among the highest in recorded history.

These events provide an opening for new capital to step in to the ILS market, helping the underlying insurance market deliver continuity in protection against future catastrophe. They also give better information on how ILS managers cope with real life losses and major events. Strategies incepted since 2013 have never experienced storms like Irma, the longest running Category 5 hurricane on record, or Harvey, which dropped the highest ever rainfall. Notably, industry models have anticipated similar events, with adjustments factoring in climate change. Though insurance-linked investment is not for everyone, 2017 is giving investors valuable additional insight on which strategies and managers are equipped to deliver the diversification and yield benefits this asset class provides.

Christine Farquhar is a Managing Director on the Cambridge Associates Global Investment Research team. Mark Wilgar, Associate Investment Director, also contributed to this article.

Originally published on October 10, 2017