Wildfire risk models struggle to predict LA-style fires

(Bloomberg) — Among natural disasters, forest fires are the great unknown. Even the most advanced scientific models fail to identify all risk factors or properly take into account the increasing impact of climate change.

It's a major problem for homeowners, insurers and investors, as evidenced by this month's Los Angeles fires, which local authorities say burned more than 14,000 buildings, including several thousand homes.

Although the risks of wildfires to the area were well known, what surprised almost everyone was the ferocity of the disaster. That's when specialist analysts must fill the gap by finding new ways to anticipate the threats of an increasingly restless world.

Wildfire risk is notoriously difficult to predict due to factors such as rising temperature levels, as well as varying vegetation, wind speeds and topographies. It is also one of the few natural disasters where human intervention – the use of fire retardants, for example – can materially change the outcome.

“If an area has very little or no inherent risk, you don't need many variables to assess risk levels, and the models tend to agree on that,” said Tammy Nichols Schwartz, senior director of analytics at Guidewire Software Inc., a insurance solutions provider. As the perceived threat in an area increases, “the accuracy of the models can vary greatly.”

Moody's RMS Event Response estimates that insured losses from the Los Angeles wildfires will be between $20 billion and $30 billion. That's on top of the $79 billion, or 60% of the $132 billion in total wildfire losses, that insurers have paid out worldwide over the past decade, according to Munich Re.

Wildfires are “a complicated hazard to model,” said Julia Borman, an industry expert at Verisk Analytics Inc., who works with the insurance industry on catastrophe modeling. What makes the process particularly challenging are the homes and buildings that “you're trying to protect are often the fuel for the danger,” she said.

In high-risk areas, models work best when there is a large amount of detailed data. How close are houses to each other? Is there a “defensible space” between a structure and the environment so that the fire brigade can safely defend a structure? Are there vents that allow windblown embers to enter?

Global warming adds an extra layer of complexity when predicting the frequency and intensity of wildfires. Models need to be updated “every two years because the climate is changing so quickly,” says Daniel Ward, director of model development at Karen Clark & ​​Co.

California, where wildfire risk levels are particularly high, recently announced plans to build the nation's first “public wildfire risk model” with the aim of improving loss predictions and helping insurers set fair and accurate insurance rates. Verisk said earlier this month that it was the first to request a review of its wildfire model from the California Department of Insurance.

According to Guidewire's Schwartz, two wildfire risk models dominated from 1997 to 2020. They all used just three variables and the results didn't always match, she said.

Current risk assessment tools, including one developed by Guidewire, include many more variables, such as wildfire history, fire suppression ability and maximum annual temperature, Schwartz said.

It's not always enough. Guidewire's 2023 model included an assessment of high winds, but not hurricane force. The devastating fires that occurred in Hawaii that year provided a clear lesson. That fire revealed that winds from an offshore hurricane played a major role in reigniting fires that destroyed the city of Lahaina.

“Our new wildfire model will include maximum wind speeds at each location, regardless of cause,” Schwartz said.

Still, investors are skeptical about whether risk modelers will ever be able to figure out all the variables behind fires like the one in Los Angeles.

Icosa Investments rarely invests in catastrophe bonds with significant exposure to forest fires, according to Chief Executive Officer Florian Steiger. “If you look at the models, there is a difference between the modeled losses and the economic reality,” he said.

Neuberger Berman will invest in multi-risk cat bonds issued by major insurers such as Allstate Corp., according to Managing Director Sophie Ware.

Still, Neuberger is concerned about “inadequate pricing given the known unknowns in modeling,” she said.

. stories like this are available at bloomberg.com