Characteristic Events

KCC’s Characteristic Event (CE) sampling methodology ensures that the KCC models provide complete and unbiased spatial coverage so that any exposure concentrations or “hot spots” can be readily determined, giving you confidence that no surprises are hidden in your portfolio.

Enhance Profitability and Control Catastrophe Risk

Identify Concentrations

CEs provide a visual and efficient way to identify exposure concentrations hidden in your portfolio

Avoid Surprises

CEs identify all locations where you can have surprise and outsized losses

Spot Opportunities

CEs show where you can grow without adding to your peak exposures and large loss potential

Earthquake
Hurricane

Earthquake

Earthquake CEs Capture the Potential for Future Events on Unknown Faults

Capturing the location effect is particularly important for earthquakes because many faults remain unknown and unmapped by scientists. For example, the fault that generated the last major earthquake in California—Northridge (1994)—was unknown prior to that event. Scientists handle unknown faults with background seismicity, i.e. accounting for the possibility of an earthquake anywhere. KCC scientists follow this approach and create background earthquake events on a uniform grid to make sure no geographical point where there could be an earthquake is missed in the models.

The Location Effect

Where an event occurs is the biggest driver of the resulting losses. A magnitude 7.0 event near Los Angeles will cause much higher losses than a larger magnitude 8.0 event on the southern San Andreas Fault. According to USGS, a magnitude 7.0 can occur anywhere in California. Therefore, it's critically important for an earthquake model to capture all possible locations. KCC's unique CE methodology avoids the potential sampling bias inherent in traditional earthquake modeling by calculating the geographic area each magnitude earthquake would impact and then spacing the background events on a grid to provide full and consistent geographic coverage during loss analysis.

Osaka, Japan

The 100-year CE for Japan is a background CE. The mapped CEs range in losses from less than $25 million to greater than $500 billion.

Vancouver, Canada

A 100-year CE background event occurring in Western Canada involves losses ranging from less than $1 million to greater than $400 billion.

Los Angeles, California

The above shows the 100-year CEs for the Los Angeles area. They range in losses from less than $25 million to greater than $500 billion.

Hurricane

Hurricane CEs Provide Deeper Insights Into Your Large Loss Potential

EP curves have tuned us to think that larger loss always means lower probability, but that doesn’t capture the location effect. For example, Hurricane Michael had the same probability of making landfall near Miami as Panama City, but had that occurred, the industry loss would have been ten times higher—well over $100 billion.

This location effect is captured by the CE methodology. The technique involves defining the hazard for different regions and return periods and then ensuring complete spatial coverage.

Consistent, Unbiased Spatial Coverage of Risk for the Entire US Atlantic Coast

The above graph shows the US hurricane CE profile for a sample company. The x-axis represents the landfall locations which are spaced at 10-mile increments from Texas through the Northeast. The red bars show the losses this company would experience if the 100-year hurricane made landfall at each point. The blue line show the company’s market share of the industry loss for each event.

Note that the event characteristics change along the coast to keep the hazard level the same.

For example, the 100- year event is a Category 5 storm in the Gulf, a Category 4 hurricane in the Southeast, and a Category 3 storm in the Northeast.

This means the probability is the same for every loss represented by the red bars.

Why Insurers Use CEs

Exceedance Probability (EP) curves capture the frequency and severity aspects of catastrophes but not the location effect. KCC’s CE sampling methodology ensures that the KCC models provide complete and unbiased spatial coverage so any exposure concentrations or “hot spots” can be readily determined. No geographic point will be missed with this approach, giving senior decision makers confidence that no surprises are hidden in their portfolios.

Important Questions Answered by CEs:

Where do we have exposure concentrations we don't know about?
Where can we have a surprise loss that is outsized relative to our peers?
How much is too much or not enough exposure accumulation in a geographical area?