More than other industries, insurance has been about data.

Actuaries have been basing their modeling on "big data" since long before that term was in use.

What's changed? There are more types of data available now, and this data can be accessed in near to real-time. This has had a meaningful impact on the insurance industry, and particularly in commercial habitation.

As a residential property owner, it's critical to know what type of tenant you're renting to. Yet most commercial habitational insurers still base their premiums on the physical risk rather than an occupant's behavior, which can often be a more relevant measurement when it comes to factors that contribute to losses.

Now, thanks to the industry's access to more and better data, insurers' loss ratio performances stand to see some significant improvement. Having access to near or real-time tenant data means a shift from looking at physical characteristics or other location-based data, where relative risk is not based on the actual tenants. Access to deeper and more accurate tenant data can mean that some property owners will benefit by gaining access to more affordable insurance.

Physical risk versus tenant risk

Underwriters typically conduct internet searches to look for insights and pictures about the property to help them make a better risk assessment. They may also issue a loss control inspection to review the roof, property maintenance and other potential hazards to calculate the overall risk. This analysis provides the underwriter with anecdotal information that is less reliable for pricing and underwriting and can be very costly.

This type of risk assessment mirrors the 1980s homeowners' insurance market when insurance homeowners pricing was based mostly on property characteristics and home inspections. The resident owner insurance risk was missing from this calculation, which is a significant contributor to insurance losses. As a result, insurers began to develop new risk segmentation techniques to assess the owner's insurance risk, like development of insurance scoring based on the owner's credit.

Shifting to analytics and deeper data insights makes sense. Human error or malfeasance is often the root cause of property destruction. Cooking, for instance, causes nearly half of all residential property fires, according to the National Fire Protection Association. That organization has also determined that smoking is the No. 1 cause of fire deaths in the U.S. Yet instead of looking at renters, insurers often go by rental rates, which is influenced more by a region or other factors than the actual insurance risk to that particular property.

Since a single multi-unit building can have several renters, calculating the overall risk level can be hard. Some tenants might be good insurance risks, others might be poor risks, but the reality is the combination is what accounts for actual risk.

If, for instance, 90% of renters have high insurance scores, while 10% have low scores. Is this better than having 100% of renters who have average scores? A habitational risk score can take both into account and offer a clearer assessment. TransUnion 2017 Internal Commercial Habitation Research Study found that 10% riskiest policies have loss ratios that are 50% higher than the average loss ratio portfolio and are two- to three times higher than policies scoring in the top 10%.

An alternate approach

In lieu of taking the risk of each individual renter into account, insurers wind up making blanket assumptions about a particular area. For instance, a map of south central Los Angeles that looks at commercial residential properties would reveal a map that shows properties with high habitational risk scores living next door to commercial residential properties with low habitational risk scores.  In fact, some 70% of multi-family properties in the lowest-income section of Los Angeles have a Habitational Risk score of 500 or higher, which is average, according to TransUnion.

Using this example shows that the actual insurance risk to property insurers is more complex than might be assumed. In fact, a more nuanced look at habitational scores and behavior reveals that properties in low-income areas often are indistinguishable from properties in comparatively higher-income neighborhoods.

Making better insurance underwriting decisions based on occupant data is now possible.  New technology aggregates resident data related to the occupant's ages and tenure. Taking all of this into account, an insurer can get a bigger and much clearer picture, including an aggregated risk score, average age and tenure on which to base their model.

Staying abreast of sudden changes

Another reason to take a closer look at renewal properties is that changes in risk are unpredictable. While most property occupant risks do not significantly change from year to year, those properties that do, can present a re-underwriting opportunity. With better and more reliable data, underwriters can accurately address these changes in risk a much more timely fashion that will lead to better loss ratio performance over time.

A smarter approach

Placing the focus on occupants will offer a better risk profile. Without using the occupant data, commercial habitation insurers are akin to auto insurers that would be trying to rate an auto policy without knowing the drivers.  Basing a risk assessment on the people inside the buildings, rather than the buildings themselves is helping drive significant and meaningful changes. For insurers, data focused on the most salient risk factor — tenants — is also a competitive advantage. Insurance is all about the data and this is data that insurers can't afford to ignore.

Anthony Sullins is the Property Insurance Leader at TransUnion. To reach this contributor, send email to asullin@transunion.com.

Related:

NOT FOR REPRINT

© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.