The Wildfire Underinsurance Gap

The recent Colorado wildfires have highlighted a countrywide claims issue for residential homeowners having total losses—most are far underinsured. Insurance agent commentator Bill Wilson recently published a blog on the topic: Another Cautionary Tale of Underinsurance. While there are myriad reasons for this epidemic of homes being underinsured to replacement value, Wilson’s post noted in part:

One would think that, among the parties involved in the insurance process – insureds, insurer, and if an agent was involved, the agent – someone would have questioned the large gap between purchase price and homeowners Coverage A limit.

…

The insurer is identified in the story and my first thought was that the insurer needs to reimagine their ‘You only pay for what you need’ sales pitch. Maybe, ‘We only pay for what you buy’ or ‘You only pay for what you mistakenly bought.’

I agree. Yet, most policyholders have no idea what it costs to rebuild their home back to a replacement cost. Most are not in the home construction business. Even if they know the replacement costs value at one time, the costs of construction are not stagnant. My experience with clients from wildfires in Colorado and California is that virtually everybody burned out of their home is far underinsured.

Rutgers Insurance law professor Jay Feinman has called this “The Underinsurance Gap.”1 He noted the pervasiveness of the problem:

Often policyholders have coverage but in dollar amounts that are less than the extent of actual or potential losses. Until the 1990s, guaranteed replacement cost coverage was the norm, ensuring that coverage would be available for the entire cost of rebuilding even in the case of a total loss. Now it is the exception. As a result, most homes are insured for less than the cost to rebuild in the event of a total loss, because even replacement cost coverage is subject to policy limits that are likely to be too low. Three of every five homes in America are underinsured by an average of twenty percent less than full value, according to analytics firm CoreLogic, whose software is a widely used tool for estimating replacement cost. Following the 2007 wildfires, the California Department of Insurance found that even though many homeowners bought coverage higher than the policy limit recommended by their insurer, more than half still were underinsured. A decade later the underinsurance gap was still substantial; a year after the North Bay wildfires in California, ‘66% of survey respondents … [knew] if they had enough insurance to cover the cost of repairing, replacing or rebuilding their home, reported being underinsured,’ according to a United Policyholders survey.

See also New Speed Camera Vans

Law professor Ken Klein has become an expert on the issue as a result his personal experience of being underinsured:

‘Ever since my own home burned down in 2003, I have been obsessed with understanding the mechanics causing and finding the solution to nationally pervasive underinsurance,’ says Prof. Klein. ‘Over the last 15 years I have researched the subject, taught the subject, counseled literally hundreds of underinsured disaster survivors, consulted with attorneys and regulators, interviewed insiders from all sides of the question, scoured public files that usually are ignored or unknown, and read the excellent scholarship of others from a wide swath of disciplines.’

Prof. Klein’s article is the result of his research and consultation with these parties. The article has been reviewed by a number of individuals who are deeply respected in their fields.2

Professor Klein published a legal study on this topic, Minding The Protection Gap: Resolving Unintended, Pervasive, Profound Homeowner Underinsurance.3 While the full article is worthy of study and reflection for those interested, part of his introduction points the finger at the insurance industry for having inadequate cost estimating software, which is used by insurance agents and underwriters to determine residential replacement costs:

The vast majority of American homeowners do not have adequate homeowner insurance, and almost none of them know it. Today, the systems insurers use to identify recommended adequate coverage limits make incidences of profound, unintended underinsurance ubiquitous. Understanding those systems is the key that unlocks the pervasive problem of unintended underinsurance, yet is an undertaking previously largely ignored by the academic and industry literature.

Most homeowners never lose their home, and so have no reason to know whether their insurance is adequate. Until the 1990s, many if not most homeowners had ‘guaranteed replacement coverage,’” meaning coverage to rebuild a home whatever the cost. This coverage has all but disappeared, however, and now the ubiquitous form of homeowner insurance, even if purportedly for ‘full’ replacement of the home, has a coverage limit. As a consequence, pervasive underinsurance is a predictable news story in the wake of a natural disaster. In 2003, after the Cedar Fire in San Diego, California, the California Department of Insurance found itself besieged by stories of homeowners who were shocked to find they did not have enough insurance to rebuild their homes. The same happened after catastrophic California wildfires in 2007 and 2008. The Texas Department of Insurance received large numbers of homeowner complaints regarding denials, delays, and claims handling both after the 2011 wildfires and after Hurricane Harvey in 2017. In the wake of Hurricanes Irma and Maria, the Florida Division of Banking, Insurance and Financial Regulation received ‘a higher number of insurance claimants than the division expected’ from ‘homeowners who had insurance policies that covered less than 80 percent of their property’s appraised replacement cost,’ and while the division could not give a percentage as to how many homeowners were over 20% underinsured, the number was ‘high enough to warrant an emergency order issued by [the] division.’ In the wake of Hurricane Katrina, litigation in Louisiana blossomed by homeowners who felt duped by the mistaken belief that they had sufficient insurance. The same happened in New Jersey after Hurricane Sandy.

When wildfires ravaged California in 2007, the California Department of Insurance (‘CDOI’) comprehensively studied the problem of underinsurance. The resulting 1550+ page administrative rulemaking file describes how insurers deploy software that purports to account for the likelihood of weather events causing mass loss and concomitant price surges. Yet even when a homeowner both relied on that software to calculate adequate coverage limits and bought 25%, 50%, 100% or even more additional coverage on top of the coverage the insurer and/or producer recommended, over half of homeowners were still underinsured. Despite the dramatic findings of the CDOI, the administrative record has not been analyzed in any academic literature to date. Simply put, the academic record helps confirm what until now was only inferred – that across the United States, most homeowners are materially underinsured, and are unaware of that fact. Most homeowners think they have more than adequate insurance.

The explanation for the prevalence of profound, unintended underinsurance lies with the cost estimator software insurers use to recommend coverage limits. The CDOI only briefly alluded to this software, and the academic world studying insurance appears largely unaware of it. These replacement cost estimators are at the heart of the problem. Through a combination of software design choices in the way that insurance is bought and sold, underinsurance is almost inevitable. For example, the software allows for a “shortcut” calculation rather than detailed analysis, and insurers compensate producers in ways that encourage using the shortcut. While the software can recalculate replacement costs and adequate coverage limits annually, producers are incentivized to not do so for fear of losing existing customers. The software requires time and expertise to accurately detail all construction components, but the deployment of the software usually relies on the homeowner to input data by answering a handful of questions in a few minutes. These are just some of many software features combined with incentives that routinely cause inadequate calculations of replacement costs that get worse over time. (Footnotes omitted)