The Risks of Hazard

With wildfire season officially here, the Risks of Hazard is republishing the below post from last year. The question we asked in October remains valid – why do insures not do more with wildfire analytics when it’s one of the most robustly modeled cat perils?

Wildfire has been big news in 2015. The USA and Canada are both having epic years, as long-term droughts combine with hotter and drier-than-normal temperatures in much of the West burning more acres and buildings than ever. Yet, when I talk to carriers about risk analytics, wildfire seems like an ignored peril. Why?

Historically, wildfire has been underwritten (or excluded) based on proximity to trees. Historically, that was adequate, too. But over the past 20 years, as towns and suburbs have expanded into wildland, the exposure to losses has increased exponentially. This week, AM Best (mind the subscription-wall) is reporting on Guy Carpenter’s estimated losses for the Western US in 2015 and the figures are eye-catching: $1.75 billion.

There is a lot of buzz in industry literature about how underwriting profit is increasingly an essential profit source, with investment returns on capital so low.

Earlier this week, I wrote about Goldfish Underwriting and referred to this Brink article (thanks again to authors Carolyn Kousky and Erwann Michel-Kerjan) to describe how flood risk varies within NFIP A zones. The Brink article summarizes a study of NFIP claims over the past 35 years. At first I thought it would be a standard-issue, “NFIP-is-broken” article, but it is completely different from anything else I’ve read about NFIP losses, and the authors' four conclusions are each worth a look.

There are two distinct analyses needed to underwrite property: risk assessment and accumulation. The first is the evaluation of a risk based on information specific to the risk, so that it can be accepted and rated, or rejected. The second limits the writing of new risks within a defined geographical area to prevent excessive losses from a single event. Both are important for underwriting cat risks: assessment because of the potential of losses up to the full value of a property (and beyond), and accumulation because cat events cause damage to swaths of properties at once.

For many cat perils, though, much of the data used to assess the risk is not specific enough – it is generalized and inadequate for underwriting. Underwriters that use generalized data for risk assessment and depend upon their accumulation to manage their portfolio are goldfish underwriters; they keep writing risks until they fill up their aggregation limits, like goldfish keep eating until they fill up their bowl.

One of the best talks I’ve heard on insurance innovation is The Changing Landscape of Risk presented by Robert Schimek of AIG (SVP and CEO of Americas Region). He was speaking at MarketScout’s Entrepreneurial Insurance Symposium in Dallas in December.

The fact that only the people in the room, plus 63 people on YouTube, have heard it is a shame. It is super informative, very entertaining (on the Insurance Technology Lecture scale), and (for me) it accomplished the elusive goals of a good talk: it asked questions that changed the way the audience thought of problems, and provided answers to questions the audience didn’t know they had. As a bonus for blog writers, it is a gold mine of material – I will be visiting Mr. Schimek’s material over and over.

The overall subject of the presentation was the importance of entrepreneurial innovation in insurance. He had to lay a little groundwork, because AIG’s view of innovation is likely to be different from pretty much the rest of the industry’s; i.e. 63,000 employees and they pay ~ $110M in claims every day…not quite your typical insurance company. Because of this, he is also able to bring some interesting holistic macro viewpoints to the conversation.