Is ‘basis risk’ the right question when considering parametric triggers?

Parametric triggers have been in use in catastrophe insurance and reinsurance for decades now, but still some cedents, underwriters and even ILS investors continue to feel the need to match the trigger to an insurable interest.

The first time Artemis happened across a parametric trigger was way back in late 1997, when Tokio Marine & Nichido Fire Insurance secured $100 million of parametric Tokyo, Japan earthquake reinsurance through the Parametric Re Ltd. transaction.

Since then use of the parametric trigger in the catastrophe bond market has gone through periods where it has been in favour, such as 2001 to 2003 when almost 45% of each years issuance was based on a parametric trigger. Or 2006 when 34% of almost $4.9 billion of cat bond issuance used one.

Then in recent years the indemnity trigger has taken over and dominated the market, with parametric triggers less than 10% every year, falling to a low of just 0.7% of last years record issuance.

The rise of the indemnity trigger in cat bonds has been a function of market pricing, insurance-linked investors increased sophistication and the desire of cedents to get a protection they can more easily fit into their tower.

But in 2015 parametric triggers are having a resurgence, jumping back up to 15% of the near $6.6 billion of cat bonds issued to-date and now at nearly 8% of the outstanding cat bond market.

Still, questions arise about the potential for basis risk in insurance, reinsurance or catastrophe bond protection that uses a parametric trigger. Interestingly the question arises typically with the cedent, but also in recent cases with ILS investors as well.

The recent PennUnion Re Ltd. (Series 2015-1) catastrophe bond is a prime example. It’s come to light that some investors in the ILS space declined to allocate capital to this transaction over fears that it could be triggered by events which did not cause a major insurance industry loss.

The sponsor of the PennUnion Re cat bond is Amtrak, of course, not an insurer but a corporate sponsor effectively, who was looking for a responsive source of insurance protection for losses that it suffered from storms and resulting surges above a certain severity.

So this cat bond is not covering insurance risks, instead it provides insurance to cover a corporate exposure to catastrophe risk, just the thing that the cat bond market and ILS investors are perfectly placed to do. But the potential for basis risk between the payout of this trigger and a wider insurance industry loss has stopped some people from investing in it.

Is this because ILS investors have become so used to providing and investing in indemnity, or industry loss, protection to insurance and reinsurance companies that they find it harder to understand a corporate exposure unless it closely matches an insurance industry exposure?

Or perhaps because investors find it harder to consume a parametric trigger within their ILS fund allocations, unless it can be more closely matched with the potential industry wide impact from a catastrophe event?

We’re not sure what the underlying reason for this could be, there are many possible permutations, but certainly it seems that some ILS investors are keen to see that a parametric cat bond could only be triggered by an event which causes a reasonable level of insured loss as well.

But doesn’t that detract from the whole point of a parametric trigger? A parametric cat bond for a corporate sponsor is designed to payout rapidly and provide financial liquidity based on an event occurring which the sponsor knows could be hugely impactful to its business.

So in the case of PennUnion Re, the trigger is designed so that the cat bond would payout for storms that caused Sandy type surges, impacting major transport hubs where Amtrak’s infrastructure and rolling stock would be most affected.

Clearly that’s exactly what the sponsor wants in this case, but still some investors have found it harder to justify the investment in this cat bond.

It’s perhaps a function of there not being sufficient risk capacity outstanding linked to parametric triggers in the market, meaning most ILS investors and funds are largely indemnity exposed making it harder to slot in a slice of parametric risk.

For cedents, the issue of a basis risk in parametric triggers also comes up frequently, despite the fact that many of them have been taking the last couple of years to analyse, centralise and rationalise their reinsurance or retrocession spend.

During this exercise you would have thought that reinsurance buyers would have been tasking their actuaries and analysts to assess the impacts of buying different layers of their programs with different types of triggers, but so far there doesn’t seem too much evidence of this.

Placing a slice of your reinsurance or retrocession program using index based triggers, be they parametric, industry loss or a slice of each, makes a lot of sense for large insurers and reinsurers. Structures backed by these triggers can provide much greater liquidity, particularly if closely defined in the initial contracts, which in essence is what a re/insurer needs after the biggest catastrophic events occur.

But it can be a task to actually work out how they fit, effort that needs to be spent by the re/insurer when they analyse their coverage. With many buyers having been through this effort, as part of reinsurance centralisation recently, we are beginning to hear of an increasing interest in and appreciation of parametric triggers.

Cases such as insurer AIG’s issuance of a six month $300 million parametric U.S. wind catastrophe bond this year, Compass Re II Ltd. (Series 2015-1), should help the parametric cause. The deal provided a prime example of how to access responsive reinsurance cover from the ILS market with parametric triggers and has helped to raise the triggers profile once again.

But still the questions over basis risk remain. It has to be asked whether this is the right question (for everyone) though?

For large insurers and reinsurers surely a layer of cover that pays out based on a specific event happening is a good thing? The liquidity that a well-designed parametric trigger provides protects shareholder value, enables faster recovery or claims payout and provides capital to help re/insurers recover after very large catastrophes.

By now, given the centralisation of reinsurance buying exercise is well underway, or complete at most re/insurers, there has to be the ability and sophistication internally to establish a way to fit some parametric protection into a reinsurance or retrocession program at most firms.

There is a good chance that this recent centralisation and rationalisation exercise will ultimately result in an increased uptake of parametric and also industry loss based coverage. It’s given companies a chance to take a step back and look more closely at how they buy their protection.

Recent commentary suggests that this results in less cover being bought, but actually the reduction is likely largely due to market forces and really what the exercise enables is reinsurance and retrocession to be bought more intelligently.

After the next major event the reduction in protection buying may well reverse, as re/insurers realise they need the risk capital on tap to help pay claims. That realisation could also stimulate a much greater uptake in parametric products, as they also understand why a fast payout is key.

If companies, insurers, reinsurers, corporations and even ILS fund managers, are to buy protection more intelligently, it surely means that they need to become experts at selecting what type of capital, structure and trigger sits where in their re/insurance and retro programs.

If buying protection more intelligently is the goal, then parametric surely fits somewhere in that picture. Buyers need to get good at matching the specific event scenarios that will hurt them most with the appropriate triggers, structures and sources of risk capital.

The benefits of a rapid payout from a source of risk capital designed to benefit the buyer after very specifically designated catastrophe events occur, is surely too beneficial for owners, shareholders or investors for the buyers not to look seriously at.

So is basis risk the right question when it comes to parametric triggers? Or should buyers be asking themselves whether their approach to buying protection needs a rethink?

Taking a step back to assess exactly what works best for them, what provides the most effective and responsive protection for their clients, shareholders and investors, and what would make their pain points less painful after the next loss event occurs, are all questions worth asking in the context of risk transfer trigger choice.