Magazine article Risk Management

Collaborative Self-Insurance

Magazine article Risk Management

Collaborative Self-Insurance

Article excerpt

If life were simple, your insurance would cost the same as your expected losses. If you faced a 1% chance of losing $10 million dollars, your premium would be 1% of $10 million, or $100,000. However, your insurer has to pay for salaries, underwriting, adjustment and shareholder dividends; and it has to account for the possibility you may have more than a 1% risk of loss. In the end, you may pay 2% or 3% for the same coverage.

Suppose you self-insure that same risk. Ninety-nine times out of 100 you will not have to pay anything, but that 1% of the time will cost you dearly. Your annual expected costs are precisely equal to your annual expected losses, which is fine if you can accept such a large potential loss. If not, you will need to transfer some of that risk.

Traditional catastrophic property and casualty insurance, however, remains expensive due to the quality of underwriting, the perception that catastrophic losses are on the rise, weak investment returns, large overhead and the high price of claims adjustment. Only a small portion of each premium dollar directly contributes to the expected losses an insured will suffer. Non-risk costs are responsible, for premium components that could be priced as much as a multiple of the current loss expectancy.

Settling large P&C claims has always been a challenge because policyholders and insurers invariably disagree on loss magnitude and responsibility. It took more than five years, for example, to resolve claims disputes stemming from the earthquakes that struck Guam in 1993 and Northridge, California in 1994. Settlement of the World Trade Center claims are still ongoing nearly three years later, with no speedy resolution in sight.

Solvency of insurers and reinsurers also adversely affects catastrophic P&C insurance. At least eight companies became insolvent as a direct result of underestimating Hurricane Andrew losses, while insured losses from the Northridge Earthquake exceeded the total value of earthquake premiums written nationwide prior to that event.

In hard insurance markets, companies may consider self-insurance. But all to often, self-insurance simply means "uninsured," since many companies do not bother to assess how they would pay for a catastrophic event.

All of this leaves companies to seek a middle ground between traditional insurance and self-insurance in which they transfer most of their risk but reduce the large nonrisk portions of their premiums. Collaborative self-insurance (CSI), a promising new strategy for covering catastrophic loss, does exactly that while also offering nearly immediate claims settlement and virtually no chance of insolvency.

Collaborative Self-Insurance for Beginners

At CSI's core is the collaborative, a small group of companies with geographically diverse P&C portfolios, who agree to provide mutual assistance following a catastrophic event.

Collaborative members have a contractual arrangement with each other rather than a traditional insurance relationship where a single entity is assigned risks in exchange for a premium. And, payments are made only if a claim occurs. Each member holds a "contribution" in an individual escrow account or through a letter of credit. For example, 10 companies that each have a $50 million catastrophic exposure might contribute $5 million to a collaborative. The contract describes the members' obligations, how much each will contribute to the capacity, what losses are covered and how payments are made. A collaborative is different than a captive because it is not a third party funded by capital investments. However, as a risk transfer vehicle, there may be regulatory issues that require the arrangement to conform to jurisdictional requirements.

CSI calculates claims using event, or parametric, contingency. Traditional insurance claims are paid based on adjusted losses. Lengthy and often expensive disputes can occur when a policyholder and an insurer disagree on a settlement. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.