Insuring Property Risks in a Captive

Anyone who's ever formed a captive understands two of the fundamental precepts of successful captives—relatively predictable losses with multiyear payout patterns.

If a company's (or a group of company's) losses and payout characteristics are reasonably predictable based on historical data, it is not difficult to project the degree of success a captive may experience. For example, if workers compensation losses typically close within 5 years of occurrence and the payout distribution is not skewed to the left (meaning that the majority of losses pay out in years 1 or 2), a relatively predictable (and worthwhile) investment income stream would result.

Another inherent benefit of this arrangement is that while a captive's early years are particularly risky (as are all startup companies), the fact that losses do not pay out all at once tends to counterbalance the startup risks and usually allows the captive enough time to build capital and loss reserves. This describes volatility risk: the longer the captive exists, the lower the volatility over time.

The vast majority of the world's captives insure casualty lines of insurance, due in no small part to the above relationship between predictable losses and their payout characteristics. Some casualty captives also insure a small amount of property risk, which serves to diversify the risk portfolio to some degree. While damage to property certainly can cause general liability and workers compensation losses, property risk is sufficiently noncorrelated with casualty risk to add some portfolio value to a casualty-dominated captive.

Given the imperatives of loss predictability and long-term payout trends, characteristics inherent in most third-party lines of insurance, how can we possibly insure stand-alone first-party risk in a captive? Historical loss activity such as fires holds very little predictive qualities relative to future property losses. In fact, after a fire, owners often take steps to minimize the risk of future fires, rendering the marginal predictive value of the loss almost inconsequential. Windstorm and flood risks are somewhat predictable given historical loss activity, but these predictions only have value when applied to a huge amount of property spread out over a large geographical area.

Insuring property in a captive is not for the feint of heart, but if you are willing to assume a significant amount of risk (and volatility) in the early years, a property captive can bestow significant downstream benefits. Unfortunately, the characteristics that help limit volatility in casualty captives actually produce volatility in property captives.

Except in extraordinary circumstances (the World Trade Center, for example) the majority of first-party losses settle and close in the year in which they occur. There is no reliable payout pattern on which to judge capital and loss reserve requirements. Moreover, there is no "tail" liability for which loss reserves must exist; each year a property-only captive literally starts with a fresh set of exposures unburdened by developing past losses or IBNR (incurred but not reported) losses. Of course, in a large property insurance program, small, unreported property losses often crop up in later periods, but this is the exception (at least it should be). So, from a risk/volatility perspective, property captives present two competing dynamics: losses pay out all at once but when the policy year is over, it's really over.

Unlike casualty insurance captives, property captives have no per-occurrence limit—the value of the funding equals the limit of liability. However, most property programs utilize blanket, agreed-amount policy language, and the value of the captive's funding in conjunction with the reinsurance does not usually match the TIV (total insurable values). Most all captives utilize three basic financing components—premiums, capital, and collateral. Regardless of how the premiums are developed, they must reflect, to a degree, that which the market would bear for the same risk. So what are the characteristics of a successful property captive?

• Significant property values—at least $1 billion, depending on geographic diversity
• Excellent geographic diversity, especially important for windstorm
• Adequate (but not excessive) premium funding
• The financial wherewithal to withstand one catastrophic loss in year one and not fold the company
• Comprehensive risk evaluation and underwriting
• Excellent loss prevention and control protocols
• A broad coverage form, with blanket and agreed amount language (reduces the number of coverage disputes)
• A significant limit of liability; low captive limits deter reinsurance participation and drive up the premiums of those who will play.

Remember, the first 2 to 3 years of any property captive are extremely volatile; any one major loss event could wipe out the captive's assets. But, if you survive to year 3, and you have funded the premium each year, the chances grow better and better that you will create a long-term facility for primary property insurance.