Smoke signals: pricing wildfire risk

“The areas most sensitive to an increased frequency of large fires and property damage have a lot of development in the wildland-urban interface.” Shelly Yerks, CoreLogic

· Pricing crucial to encouraging wildfire mitigation efforts
· Cat level events ‘most sensitive to climate change’
· Frequency of large wildfires set to increase in years ahead
· Pandemic lockdowns and quarantines already hindering fire suppression work

As the images of huge wildfires become an annual feature on global news reports, cat risk modelling experts think insurance pricing could encourage more fire mitigation efforts as well as offering improved protection.

In 2017 and 2018, wildfires devastated large areas of the US, shocking many with their intensity and scale.
These years are now recognised as two of the worst on record as the fires laid waste to more than 10 million acres of land in both years.

America had some respite from this type of catastrophe in 2019, described as “a relatively low year in terms of wildfire acreage” by Tom Jeffery, principal, science & analytics at insurance data firm CoreLogic.

But in 2020 and beyond we may see a repeat of wildfires engulfing forests and homes.

Until the middle of August this year, 2.4 million acres had already burned in the US, with notable events such as the Apple Fire and Carmel Fire, both in California.

With so much at stake, how can insurers price the risk to encourage more fire mitigation efforts and better protection?

Researchers at CoreLogic are looking at how to incentivise wildfire mitigation with risk-based wildfire insurance pricing.

Of course, this year has the added complication of the COVID-19 pandemic. Lockdowns and quarantines have already hampered fire suppression work, reduced the number of firefighters available and delayed recovery and rebuilding work.

There is hope that the threats posed by wildfire can be better managed and risks reduced.

Maiclaire Bolton Smith, senior leader of research and content strategy at CoreLogic, says it can be done by analysing the risk and prioritising property level mitigation.

Jeffery says that in the last five years, the US has experienced some of the worst fires in modern times. But he adds: “It goes without saying that the location of the fire is ultimately the most important factor, for example a fire in a densely populated area that can cause a lot of destruction even in a low-impact year.”

As of August 16, 2020 there were 35,878 fires covering more than 2.3 million acres in the US. These figures were expected to increase.

The 10-year annual average (2010 to 2019) shows 39,463 fires and 4.7 million acres, making this year the second lowest number of acres affected and the third lowest number of fires over the last 10 years, so far.

Pricing to incentivise mitigation

When assessing how to price wildfire risk, CoreLogic looks at what affects the expected loss for a particular structure, says Howard Kunst, chief actuary at the firm.

“In basic terms that is frequency—the exposure to the hazard. For the individual structure there are things that affect the severity of the loss: construction type, roof type, the age of the home, and fire mitigation that is in place.”

He says that CoreLogic ran a few simulations for California with different ages of homes and roof type to see what happens with frequency and average annual losses.

Using a fire risk score of 1 to 100, Kunst says the simulations showed, as might be expected, that the risk score went up as frequency increases.

Kunst says the models simulating expected average annual loss versus total insurable value show that as the fire risk increases the difference between the annual average loss between wood frame and unreinforced masonry homes rise to 15 percent at the top end of risk.

“We would expect wood frame homes to be more damageable, but here it’s showing about 15 percent relative to the construction.”

Looking at the losses for a home built in 1975 versus one built in 2015, he says that the older home is significantly more damageable than a newer one.

“There are other things that can go into our models such as mitigation of fire and clearance around the structure, so there’s less fuel around it, which reduces the opportunities for fires and losses.”

Roof type is another variable used in the model, so a less flammable material would be lower risk and enable a lower insurance price, which is not a bad incentive for wildfire mitigation.

“These are just some examples of the possible mitigation and how they affect the overall losses and the prices that you get for writing this type of business.”

Climate change sensitivity

Shelly Yerks, senior professional product management at the firm, says climate is also impacting wildfire but it’s complicated because there are a lot of dynamic components that can change from region to region. But she is clear that one of the effects of climate change will be larger and more destructive fires.

To assess the impact of these large wildfires on the number of homes burned and the damage done, the firm built a model using California data. It revealed that while large fires, affecting more than 100,000 acres, are less frequent, making up just 4 percent of the total fires, they cause 35 percent of the damage in the state.

“These cat level events drive a disproportionate amount of the loss. These are the fires that we found are sensitive to climate change. And we believe there’s going to be a higher frequency of these large fires in the coming years,” says Yerks.

There are also particular areas that insurers need to be more concerned about. Using the northern region of California as an example, Yerks explains that this is where the Camp Fire burned down the town of Paradise in 2018.

“In this region, there are a lot of communities in the wildland-urban interface and many of them don’t have a lot of good access for effective fire suppression.”

Within these regions, Yerks says, the areas most sensitive to an increased frequency of large fires and property damage have a lot of development in the wildland-urban interface. They have a bigger population to potentially start more fires (80 percent are started by people) and there is more wind to keep fires going.

“These are the areas you need to be concerned about and mitigate your risk because they’re going to be the most sensitive to the increasing frequency of large events that are going to come with climate change.

“I use the term ‘mitigate’ because pricing your way out of the problem is not going to be the only answer. Mitigation efforts such as hardening the roof or the property perimeter can have a big impact on the vulnerability of the structure.

“In the most at-risk areas community-wide mitigation programmes are going to be necessary, in addition to mitigation of individual properties.”

Wildfires are part of Californian, and US, history and that is unlikely to change. Bolton Smith says that with the planet’s evolving climate, society must adapt and continually improve how it manages the risk.

“We know we cannot control the occurrence of a devastating fire, but we do have the data and analytics to help us understand what can happen, and better knowledge on what can happen will help us achieve a more resilient future,” she concludes.