How data analysis quantifies water, fire risks for insurers
We can't always blame Mother Nature. Now, insurers can quantify water and fire risk from non-weather-related events.
The weather is unpredictable. And yet, sophisticated catastrophic risk and weather verification models have made it possible for property insurers to quantify the ‘what could, what if and what did happen’ of natural hazard risk.
Surprisingly, outside of storm surge, wildfire, winter storms and other weather-related events, tools to help quantify the risk of water and fire damage from non-weather-related events have been nonexistent.
So how can property insurers close the gap on pricing risk associated with damage that isn’t caused by Mother Nature?
The risk of sparking a fire
One of the most common ways the industry assesses and prices fire risk is by evaluating the severity of the exposure to a property once a fire starts. Components such as distance to a fire station, fire station attributes and distance to a fire hydrant are typical inputs for standard-risk assessment models.
However, with the proliferation of synthetic petroleum-based composites in construction materials and home furnishings, structures burn much faster today than 30 years ago. In fact, studies show that today’s homes burn nearly six times as fast. Therefore, assessing the severity of fire risk has become less predictive of insurance losses than knowing the risk of a fire starting in the first place.
According to the National Fire Protection Association, 84% of home structure fires result from malfunctions of either cooling, heating, electrical and lighting systems or fixtures, are intentionally set, or involve smoking, candles, or cooking accidents. In fact, lightning and wildfires cause only 16% of fire damage to structures.
Related: Top 10 ZIP codes for fire loss
Plumbing failures: A leading source of water losses
According to a Location, Inc. analysis of insurance carrier loss data, non-weather-related water damage accounts for approximately 20% of all homeowner insurance losses nationwide. Given that a single weather-related event, such as a hurricane, can impact hundreds of thousands of properties in one swoop, this statistic is indicative of the true problem of non-weather water-related mishaps that can impact a carrier’s bottom line.
However, water damage risk models that score the risk of potential damage to a structure from events such as appliance failures, air conditioning leaks and even overflowing bathtubs were previously unavailable to the insurance industry — until now.
It’s all in the science
A successful solution to the problem of non-water and non-fire-related loss must have nationwide coverage, be localized and extremely granular and contain millions of records of practical, measurable data.
These risk models should assess the interaction of all elements to predict the probability and magnitude of potential damage to a structure from fire and water mishaps.
Because as Murphy’s Law states, “Anything that can go wrong, will go wrong.”
Steve Brewer (stbrewer@corelogic.com) Steven Brewer leads the Insurance and Spatial Solutions group within the Property Information and Analytics division of CoreLogic®. He is responsible for the creation and execution of overall vision, thought leadership, planning, execution and growth of the Insurance and Spatial Solutions business.
The opinions expressed here are the author’s own.
See also:
Why insurers, reinsurers should care about silent risk exposures
The risk modeling benefits of downward counterfactual analysis