The COVID-19 pandemic has had a profound impact on almost all facets of civic and corporate life—from employment and commerce to how vigorously we wash our hands.
One of the most visible imprints of our emerging “new normal” has been the fluctuating levels of traffic, particularly during March and April. For personal and commercial auto insurers, the pandemic and ensuing lockdowns have led to a potentially dramatic change in the risk environment for vehicles.
The use of experience data (i.e., historical loss costs) alone may not always provide the necessary analytical flexibility; however, it’s usefully complemented with predictive analytics to gauge the potential impact of this disruptive new normal environment on pure premiums.
A more localized view of environmental risk
What can our predictive models tell us about how future loss costs may develop? We utilized ISO Risk Analyzer®, Verisk’s suite of predictive models that helps classify, segment, and price insurance risks, to perform a stress test examining how pure premiums would respond to several economic scenarios. These models, which are available for personal and commercial auto, homeowners and businessowners, can help insurers predict expected losses at a highly refined level according to each risk’s unique characteristics. The Environmental module in ISO Risk Analyzer Personal and Commercial Auto examines the interactive effects of hundreds of variables to provide more refined estimates of loss costs by geography.
For this exercise, we used two components of the module—Traffic Density and Traffic Generators—to derive a traffic risk score attached to a given geography. We then seeded our model with an economic variable from ISO MarketStance Commercial Insight, which provides estimates for various economic and exposure measures, including a firm's sales, payroll, and number of employees. We selected business operating locations as our variable under the hypothesis that among the main reasons vehicles are on the road is to commute to work, frequent a business, or make deliveries. As businesses close, we can reasonably expect a reduction in traffic to those locations and some of the surrounding areas.
For our stress test, we posited four economic scenarios: a 5, 10, 15 or 20 percent reduction in business operating locations. ISO Risk Analyzer was then used to estimate how much the effects of reduced mobility to these shuttered business locations diffuse as you relate them back to insurance risk. The effects of these business closures can be subtle when analyzed for traffic risk in auto insurance. Think of it like this: if two of the four local pizza shops close in a given ZIP Code close, the remaining two may re-absorb some of that business, keeping vehicles on the road. And the vehicles traveling for reasons beyond an unquenchable desire for pizza would be unaffected by those closures. But it’s worth remembering that a reduction in traffic risk is only one piece of the puzzle in auto insurance. Additionally, the impact seen can be amplified across coverages and lines of business.
Searching for connections: butterfly wings and tornados
In chaos theory, the “butterfly effect” suggests that small changes can propagate over time into much larger disruptions (in this analogy, the flapping of a butterfly’s wings ultimately leads to a tornado). There’s a similar dynamic at play with insurance premiums—small changes to pure premium on the individual policy level can, when aggregated across a large book of business (let alone an entire industry), lead to larger swings.
As we analyzed the results of our stress test across personal and commercial auto pure premiums in four coverage areas—Bodily Injury, Property Damage, Collision, and Comprehensive—we definitely had the butterfly effect in mind. Some scenarios produced more dramatic “wing swings” than others (and, perhaps counterintuitively, not always negative swings) and all reinforced the need for powerful data assets and analytics to more successfully excel in the dynamic times ahead.
Take a deeper dive into our analysis to understand how to harness predictive power to better navigate the new normal, and download the latest white paper, “Harnessing predictive analytics to gain pricing insights in uncertain times.”