Competing in the insurance market through differentiation, and demonstrating knowledge and expertise to a client, are central to so many business strategies in this industry. The client values the insight an insurance business delivers on their exposure which is reflected in their premium. Sometimes, taking the regular model output view of risk is exactly what’s called for. But to demonstrate this differentiated offer, what about a view of risk for a specific class of buildings, or even just one building?
By contributing your own vulnerability data into a model, you can deliver a custom view of risk that can get you to a new granular level, whether it is for a specific twenty-story apartment block on Miami Beach or a hundred “big-box” stores that share similar characteristics. With this specialist view, you can differentiate your offer in the eyes of your customer, and generate a “win-win” situation of better downstream pricing for the insured, and profit opportunities for your business.
For many insurers, reinsurers and brokers, there is no shortage of custom vulnerability data. Data is collected from risk assessment and loss control inspections, either using their own or third-party inspectors. Some of these inspections that focus on properties in the top ten percent of a business’s portfolio can cost up to $5,000 or even $10,000 per inspection and run to five hundred data points per building.
What happens to this data? The inspector will compile a risk engineering report, which will be passed to the underwriter, who is then responsible for acting on this data. Much of this can end up as a manual workaround, outside of the modeling or general business workflow. A busy underwriter who may not totally understand the implications of the report may overlook important aspects contained within the findings.
Without the data being included into modeling and into a subsequent Exposure Data Model (EDM), or the right conclusions drawn, the substantial investment made in loss control inspections and reporting will be unrealized. This data needs to be embedded into the workflow and used within the modeling process, which requires the ability for the captured data to be entered for modeling and for customized output to be generated – using an approach called open vulnerability modeling.
Open Vulnerability Modeling
Open vulnerability modeling is exactly as described, RMS allows users of its RMS RiskAssessor™ custom vulnerability solution to integrate their own distinct view of vulnerability into their modeling workflows. Using RiskAssessor means realizing the investment your business is making in collecting custom vulnerability data, acting as a tool to collect, and centralize all the data, together with the opportunity to expand the range of variables you are collecting data for. And rather than sitting outside of your modeling workflow, it is integrated within it, to help continually refine your view of risk.
RiskAssessor currently includes the ability to generate custom vulnerability relationships for the U.S. hurricane wind peril, and offers almost quadruple the number of variables compared to the inclusion of standard hurricane model secondary modifiers. These variables all represent data capture points that anyone from a site inspector to an engineer or a catastrophe modeler can understand and use to effectively represent a structure. Using RiskAssessor, a user can now capture the fact that a building has 90 percent coverage of glazed openings for instance, a vital factor for assessing hurricane risk.
The majority of the variables available in RiskAssessor are unique compared to RiskLink or allow for additional detail. The variables look at building geometry, with the explicit representation of different plan shapes and additional corners that change wind load distributions. It can take into account the difference between varying quality and rating levels for individual components such as the roof, cladding or glazing, and represent the glazed opening area coverage on a building.
Set-up time to customize a view of risk and implement it in the model can take as little as five minutes, and a custom vulnerability curve can be easily compared with the standard curve in the modeling workflow. Aided by RMS inventory research that provides typical building characteristics for a given combination of region and primary characteristics, a user can quickly start to unlock credible risk differentiation across their portfolio.
The effect of including custom vulnerability data on a customer’s premium and on profitability can be dramatic. We have seen users who have spent five minutes using public mapping sources to collect basic assessment data, reduce the risk of an individual location by 50 percent or more, resulting in significant cuts to a customer’s premium.
Although RiskAssessor allows for a per-building approach, generating a customized EDM for a class of properties, such as “big-box” stores sharing similar building characteristics, offers the opportunity to demonstrate your differentiated approach. Once created, these vulnerability curves can be re-used over and over again for any EDM, and genuinely represent your unique view of risk for a class of buildings across your portfolio.
What could this approach mean for your business? Some RiskAssessor users are moving to specialize in types of buildings where they know that using a detailed representation of vulnerability will make a significant difference compared to the standardized model output, and can credibly and confidently expand their business. To find out more, watch our short RiskAssessor video, or download our datasheet.
You May Also Like
January 22, 2018
Southern California: Thomas Fire Loss Estimate and Mudslide Commentary
California Wildfires: Exposure Impacted by the Thomas Fire
As the Thomas Fire continues to climb the list of the top twenty largest California wildfires for both acres burned and structures destroyed, many in the insurance industry are asking how this fire, in addition to the other burned areas across Southern California, will impact their portfolio. A critical element in understanding the industry impact, but also the significance for an individual book, is the insured value of the burned structures. The Thomas Fire, which at 60 percent containment at the time of publication is already the second largest fire in California history with a reported burn area of 272,000 acres (110,074 hectares), has affected several different communities with wide ranges of average insured value.
Figure 1: Top 20 Largest California Wildfires. Source: CAL FIREWhile damage assessments are still ongoing and counts of damaged or destroyed structures are actively being reported, we can use RMS high resolution exposure data and the latest burn footprint available from Geospatial Multi-Agency Coordination (GeoMAC) to gain a perspective on the total amount and average value of exposure across different areas within the perimeter.
Figure 2: Map of the Thomas Fire Perimeter. Source: GeoMAC as of 0100 UTC on December 18, 2017.As of December 18, portions of ten different ZIP codes are located within the Thomas Fire perimeter, spanning across Santa Barbara and Ventura counties. While the fire started north of Santa Paula around sunset on December 4, it has now spread approximately 50 miles (80 kilometers), primarily to the west, to the hills north of Montecito and Santa Barbara. So far, during two weeks of spread the fire has impacted the communities of Santa Paula, Ojai, Ventura, Oak View, Wheeler Springs, Carpinteria, Montecito, and several others. Each of these communities vary in average residential exposure value, making “average” estimates across the entire perimeter more difficult.
Table 1 below compares the residential exposure for the high value 93108 ZIP postal code that spans the Montecito area against the exposure outside of that ZIP code. All values correspond only to exposure located within the Thomas Fire perimeter, according to the RMS high resolution exposure database. The average residential structure exposure value for 93108 is on average three and a half times higher than structures located outside of that ZIP code. Structures in 93108 accounts for only three and a half percent of the number of structures within the fire perimeter, but 12 percent of the total exposure value.
Table 1: Residential Exposure Values in the Thomas Fire Perimeter.As noted in prior blogs about the Wine Country wildfires, it is important to note that only a fraction of the structures within the perimeter will be damaged or destroyed in this event. The reported numbers so far from CAL FIRE are 1,024 structures destroyed and 250 damaged. Comparing this against the total number of structures within the perimeters, the ratio is approximately 37 percent, which is far lower than the 75 to 80 percent seen in the Wine Country wildfires in October this year, but higher than the other historical Southern California events that have been analyzed (Cedar in 2003 and Witch in 2007).
Figure 3: Percentage of damaged or destroyed structures within historical fire footprints.Also, a comparison of the percentage of destroyed structures versus the total number in the perimeter between the Thomas Fire (1,024 destroyed versus 1,274 total or 80 percent) and the Wine Country Tubbs Fire (6,957 destroyed versus 7,443 total or 93 percent), raises important questions for understanding the key differences between the events. This may indicate a stronger presence of fire suppression, lower impacts of embers, more distributed exposure, differences in surface fuel characteristics (e.g. chaparral versus forest) or a combination of these and other factors in the Thomas Fire. A further review of these event-specific factors, including the weather conditions, will provide more clarity around key drivers of these differences.
By extracting information about the differences between these two events, RMS will continue to build insight into the development of the RMS® U.S. Wildfire HD model, part of the RMS North America Wildfire HD Models suite, due for release in 2018. RMS is still monitoring the ongoing events affecting Southern California and will continue to provide updates through RMS Owl.…
As a senior product manager in the Model Product Management group at RMS, Kevin is responsible for RMS climate-peril products for the Americas, including wildfire and custom vulnerability analytics. Kevin has been actively involved in model releases for both severe convective storm and hurricane models over the last four years at RMS. Kevin holds a master’s degree in atmospheric science from the University of Illinois at Urbana-Champaign, where he authored a thesis on tornado-genesis and severe convective storms, and a bachelor’s degree in atmospheric science from Purdue University. He also holds the Certified Catastrophe Risk Analyst (CCRA) designation from RMS. Kevin is a member of the American Meteorological Society (AMS), a mentor for the AMS Board of Private Sector Meteorologists, and a voting member of the ASCE Standards Committee on Wind Speed Estimation in Tornadoes.