The Intelligent Quarterly from the publishers of The Insurance Insider

Winter 2011 / 2012

Search archive

Advanced search

A seismic analyst at the Caltech Seismological Laboratory shows the 7.0 earthquake peak from the Haiti earthquake One of the most basic tenets of insurance is that policyholders must have an "insurable interest" in order to buy cover protecting their assets. But is the industry on the cusp of moving away from this principle?

Marsh CEO and Insurance Institute of London (IIL) president Martin South argued in a recent IIL speech that the industry is wedded to the "outdated concept" of insurable interest. South said the industry was not delivering on its promise to pay and that it should expand its use of "pay now, argue later policies" with "smarter triggers". Parametric and other derivative solutions at least "do what they say on the tin", he said.

Unlike traditional indemnity (re)insurance that covers a policyholder's actual losses, parametric insurance pays out in response to defined triggers. This means the insurer can quickly establish how much it owes the policyholder - providing quicker, smoother pay-outs. Parametric deals can be settled in days or weeks, compared to the weeks, months or years it may take to finally close a traditional (re)insurance payment.

Parametric insurance evolved in the late 1980s and early 1990s in the nascent catastrophe bond market, where it is still predominantly used for natural catastrophe risk. More recently, parametric solutions are increasingly being deployed in public sector and developing markets, as well as being bought by corporate clients.

The speedy pay-outs suit both capital market investors providing cat bond cover and public sector buyers that may need cash-flow after a disaster. Investors prefer not only the short-term horizon but also the fact they can more easily gauge the odds of paying out parametric insurance.

Parametric cover is also cheaper than indemnity (re)insurance, but the lower prices reflect the greater risk that a policyholder may not recoup their true losses from a disaster - a threat known in the industry as basis risk.

Basis risk
Basis risk is the main stumbling block to further development of the parametric insurance market, admits Andreas Schraft, Swiss Re's head of natural hazards, who argues that, nonetheless, there is a lot of potential for the market to grow. "To make that growth reality, we will need to come up with solutions to address basis risk," he says.

Click to enlarge Swiss Re deployed one such new structure in its MiCRO programme in Haiti, which insures microfinance loans against natural disasters. Instead of the policyholder taking on basis risk, a captive company funded by donors and the World Bank bears this risk.

Other examples occur in the cat bond market, where reinsurers enable clients to access parametric insurance without taking on basis risk by writing them a traditional indemnity policy and then passing off some of this risk to the capital markets through catastrophe bond issuances.

Munich Re did this with the Muteki bond for Japanese mutual insurer Zenkyoren, which recouped the full $300mn face value of the bond from investors within two and a half months of the March Tohoku earthquake.

But Munich Re's head of risk trading, Rupert Flatscher, reiterated at a recent legal conference in London that there is still a marked difference between cat bond investors' preference for parametric deals and potential cat bond issuers' reluctance to take on the basis risk.

Parametric transactions are generally only suitable for very large companies, because they are harder to tailor to more idiosyncratic portfolios where basis risk can be enormous, he said. However, Schraft argues that willingness to take on basis risk comes down to the level of comfort with catastrophe risk and the understanding of risks.

So-called hybrid deals - which mix the attributes of indemnity and parametric cover - are emerging as a popular solution to the cat bond market's dilemma. These solutions adapt industry loss triggers - which pay out based on overall industry losses as measured by third parties - to suit a particular customer's book. This can be done by weighting the industry loss trigger to put more emphasis on certain zones or to break losses down at a lower level of industry loss.

Less wriggle room
So does parametric insurance truly deliver "what it says on the tin"? It could well be a way of getting insurers out of the courtroom once and for all. Experts agree that there is little wriggle room in parametric cover to allow insurers to escape paying claims.

"Experts agree that there is little wriggle room in parametric cover to allow insurers to escape paying claims"

Michael Madigan, partner at law firm Sidley Austin, says that from a legal perspective there should be less risk of a dispute with investors when a parametric or industry loss trigger is used where payment is based on calculations by third parties. "There isn't much to argue about unless there's been a pure miscalculation," he says.

Schraft agrees that there is not much room for dispute as long as triggers are well defined and safeguards are built in. And, like indemnity deals, the risk of parametric triggers not performing as expected due to modelling errors or unknown elements of catastrophe risk has always been there, says Madigan.

For example, while Muteki paid out after the Tohoku earthquake, other Japanese-exposed cat bonds were untouched despite the magnitude of the disaster. In some cases, this was because they were designed to protect against the peak insurance risk scenario of damage concentrated in Tokyo. Risk models did not capture a scenario involving a quake of such magnitude followed by a tsunami.

Indemnity reinsurance may also involve basis risk for the purchaser if they choose to build in specific limits on what is covered by the policies - which may mean the cover is not a perfect match for their losses, even while it will minimise potential disputes. "The more basis risk, the less risk of disputes," Madigan says.

Schraft predicts that in the short term, the most growth in parametric cover will come from the corporate and government market. But over the long term, he sees greater potential in developing countries. Here, parametric insurance helps reinsurers access the market when countries lack the infrastructure to distribute and support their products.

And Schraft says that while the cat bond market will remain the biggest source of parametric cover in the short term it will eventually be overtaken by traditional providers. "It's a new way to provide cover to get to clients we couldn't reach previously, to insure risks we couldn't insure otherwise."

New horizons
The parametric market is currently testing out new perils including business interruption (BI), contingent BI and energy contracts. Munich Re's proposed $10bn SOS facility for offshore energy clients - which it drew up after the Deepwater Horizon/Macondo oil spill last year - would use a parametric trigger based on how many direct compensation claims had been made against the insured.

Meanwhile, Cooper Gay's Latin America CEO Steve Jackson recently cited the potential to develop parametric drought insurance based on vegetation indices as one of the ideas at the "cutting edge" of such contracts.

Industry loss aggregator ISO has also recently released a new casualty index, which has been licensed by broker Jardine Lloyd Thompson. ISO says the data can be used as a benchmark to provide accident year stop-loss and adverse loss development covers.

However, Flatscher says cat bond investors have not yet proved to be of much assistance to (re)insurers in taking on casualty risk. "If they want to accept casualty risk with a longer time horizon, the risk could be transferred," he says.

The only previous casualty cat bond issuance, the 2005 Avalon Re deal for Oil Casualty Insurance (OCIL), was an indemnity deal covering third-party liability claims. The $405mn transaction finally paid out $13mn of losses on its delayed maturity in 2010, two years after its planned expiry.