Property Insurance Woes...

Property insurance markets are unsettled. The industry is struggling with rising loss dollars, reinsurance capacity concerns and emerging exposures that challenge the pricing and risk selection models in place. The property insurance business faces these forward-looking challenges from a perspective of looking back, not forward. This makes managing today’s risks more difficult. I will focus on areas where the look back approach fails to deliver guidance.

Fire Behavior is making past loss experience irrelevant

The National Board of Fire Underwriters, was formed in 1866, it was in place before the NFPA.. The first action of this group was to initiate a data collection process around fire losses. This was the foundation of the classification and rating systems used today.

This was the first “big data” undertaking that would allow underwriters to analyze past experience and project what rates would be needed in the future.

As the name suggests, the focus was on the risk of fire. This singular focus has persisted in property underwriting since 1866. A driving focus was conflagrations that destroyed large portions of cities in a single event. This led to the development of many common underwriting tools. The Public Protection Class (PPC) was a product of this group.

Most property insurance carriers have clear space charts for defining fire divisions. This is based on fire spread. Loss projection tools such as Probable Maximum Loss (PML) are routinely based on a fire as the event modeled. We look back at prior loss history to validate these tools and the results they develop.

The weakness of looking back to the 1800’s for benchmarking is that there has been a shift in the built environment that makes data from before 1990 obsolete. I base this on my experience as a fire chief as well as an HPR property consultant.

I ask that you look at the video of fire tests performed on dwelling contents conducted by Underwriters Laboratories (UL). These tests compared the fire growth of current furnishings to those typically found 25 years ago, legacy furnishings. The video is available on YouTube https://youtu.be/87hAnxuh1g8

Flashover occurs in modern furnishings in less than five minutes. This compares to twenty-five minutes for legacy furnishings. My time in the fire service was from 1978 to 1998. During this time, we had the benefit of the 25-minute flashover scenario, and we could respond as a volunteer fire department and save homes with aggressive interior firefighting operations. This is the type of loss experience that fire data before 1990 reflects.

Today the structure is in flashover while the fire department is either still in the station or on the road to the address. Interior operations are untenable and exterior fire stream application is the norm. This change has converted most contents fires from partial losses to constructive total losses. The fire department is now focusing on preventing fire spread to the next structure. The financial impact of this is diluted when the database includes events from the 1800’s to current time. The look in the rearview mirror does not highlight the shift in experience.

Since the current typical furnishings are everywhere, this same shift in fire behavior translates to manufacturing, warehousing, and any built environment. The aggressive nature of fire behavior should make most property underwriters rethink their approach, including reliance on Public Protection Class (PPC) as a significant tool in rating and risk selection.

The fire experience of today is not analogous to the past. The continued use of these benchmarks prevents property underwriters from making good choices. Instead of understanding the behavior of fire today, the industry forces a best fit curve on their pricing tools. The number of fires is lower, but the cost is higher.

Stay tuned for the second part of this discussion, how building construction has changed the behavior of buildings during a fire.