Do catastrophe models understand insurance?

Catastrophe models are complex but not nearly as complex as reality. A simplistic description of insurance policies limits the accuracy of risk analyses

Robert MedeirosThe risk of an insurance portfolio is calculated by taking into account dozens of parameters for thousands of insurance policies, but unquantifiable factors can have a huge impact on actual losses paid by insurance companies. In 2014, 72% of new cat bonds were designed to be triggered by the actual losses paid by insurance companies. As the insurance market becomes more competitive, policies are becoming more generous and it is more important than ever to understand what the models miss.

Robert Medeiros from Lighthouse Consulting has been highlighting some of the issues that can cause a gap between models and reality. In his first article he reviewed the realities of the claims environment in Florida. In the second of three articles, he reviews features of insurance contracts that are poorly modelled and could increase losses to insurance companies and investors.

Insurance policies are complex. My homeowner’s policy is 25 pages, plus an additional 16 pages of endorsements. My renewal came with 5 additional pages to explain the changes in coverages. And homeowners' policies are not as complex as commercial policies. A large commercial insured can have a policy that is easily 70 pages. How do cat models account for such complexity? They really can’t. The models are blind to policy coverages but there is no question that broad coverage increases the loss.

What are the coverages that are a concern?

  • Any coverage that could increase the cost to rebuild the structure. Policies usually have some limits for increased cost of construction, or increased cost to upgrade due to building codes;
  • Any coverage that provides loss of income or covers ongoing expenses. These are notorious difficult to estimate, and certain coverages can allow the insured to collect even after the damage is repaired;
  • Any coverage or exposure that could increase the loss from the inside of the structure. These typically include replacement cost on contents, high quality of interior finish, and scheduled property such as fine arts, jewelry, or antiques.

Commercial lines deserves a special mention. Policy terms can be very broad and limits very high in the large commercial segment. Not only is flood typically included but difficult to model coverages such as business income, extra expense and contingent business income are usually included. The policies also has high limits for unscheduled locations, which by definition cannot be modelled.

Finally it is worth noting that the Florida homeowner’s market has become more competitive. Rates have increased, reinsurance costs are down, and Citizens’ is actively de-populating. We expect to see insurers competing on coverage as well as price, which will impact the actual loss. Screened enclosures come to mind. Long the subject of coverage restrictions we see some companies making higher limits available.

The take away message for investors? First, ensure that you know the profile of the portfolio. A portfolio of homogenous personal lines policies should have more restrictive policy language than commercial lines. A mixed commercial/personal portfolio needs more in depth review, possibly by having separate modelling analysis for personal and commercial. Second, for a purely personal lines portfolio, ask the broker for benchmarking statistics to allow a comparison against other portfolios. Key statistics include:

  • Dates of construction-older homes are more susceptible to increased cost of construction due to advances in building codes;
  • Percentage of high value homes - this is a desirable market segment and insurers offer broader coverages. These home are more likely to have the types of high valued contents that will increase the loss;
  • Percentage of dwelling fire (DF) policies-these policies are typically used for rental properties and contain fewer coverages than a typical homeowner’s policy;

For all their sophistication, models are only able to capture a very simplified version of the terms and conditions in typical insurance policies.

Robert Medeiros CPCU, ARe, AMIM, ASLI is the founder and President of Lighthouse Consulting, LLC,, which provides underwriting audit and consulting services to the off-shore reinsurance market. He can be contacted at

Posted: Monday, January 19th, 2015