In the bottom left box are low-frequency, low-severity risks that are infrequent and uncommon, he said. An example would be “kids in a park who spray-paint buildings every so often.” The graffiti would be painted over, and the cost would “probably not even be funded for. It's just part of your operational expenses.”

The bottom right box contains high-frequency, low-severity risks. “This is the typical area where you want to self-insure,” he said.

The idea here is “not to trade dollars with the insurance company. You are taking on more risk, but the rewards should be that in a large portion of your program, you are not going to pay insurance company profits and overhead.” This box may contain “a lot of claims, but they don't tend to be large,” he said.

In the top left box are low-frequency, high-severity claims, Mr. Wightman explained. “This is what insurance companies are for–your fires and your big trip-and-falls and large liability cases, where it makes sense to transfer that risk to the insurance community.”

Insurers, he said, are in the business of taking risk and pooling it, “and hopefully charge you a reasonable dollar to participate in that sharing.”

The top right box contains risks that are high-frequency and high-severity. With these risks, “you're in big trouble,” he said. “Those are the things you would probably like to insure but the insurance community isn't interested.”

Examples of this type of risk are the California earthquake and flood plains. “A lot of entities in Southern California either can't find quake or don't buy quake [coverage], or it's so expensive it's not worth buying,” he said. If coverage is not available at a reasonable cost, “you just have to figure out some way to set up reserves for those or be ready for the consequences,” he added.

Any building located on a flood plain should be covered by the National Flood Insurance Program, he said, which is there “because insurance companies don't want to cover that.”

Mr. Wightman recommended self-insurance when possible, “since nobody wants to trade dollars with insurers.” But self-insurance should be carefully considered, he warned.

Mr. Wightman referred to a matrix that he developed to help risk managers compare several different forms of risk-financing alternatives.

Categories for comparison include program description, expense stability, security concerns, local control, administrative considerations, fixed costs, program limitations, who holds the money, and current popularity of the program. The final column rates the program, from one-to-five stars.

A one-star risk-transfer program, he said, indicates that “you're paying a premium and someone else will pay your losses.” Examples of one-star programs include first-dollar insurance, self-insurance pooling, risk retention groups, group rent-a-captives, and group captives.

Two stars indicate a moderate risk transfer with some small deductibles. The only two-star alternative program listed is the moderate deductible program.

Three stars mean the program is “at risk for routing losses, but there is some aggregate protection.” Three-star risk financing alternatives include incurred loss retro, paid loss retro, all-lines aggregate, single-entity rent-a-captive, single-entity captive, multiline/multi-year, and per-line aggregate.

Four stars indicate the method is at risk for routing losses without aggregate protection, and “you're taking on more risk, but the premium is not as much.” The only four star listing is specific-only retention.

Five stars indicate that you are at risk for all losses. “There are a couple of different risk-financing alternatives to deal with that,” he explained. Listed under five stars are finite risk and funded self-insurance.

Last on the list is “Whoops!”–in which case “you forgot to buy insurance or you didn't realize you had an exposure,” he said. An example would be some environmental exposures.

Mr. Wightman said that risk retention groups–started in 1986 when a federal law was passed that allowed for-profit or not-for-profit entities to pool together across state lines for liability insurance–have not been popular for public entities.

“There is a well-run RRG for higher education,” he added, but the limitation is that only liability can be covered in the risk retention groups.

Group rent-a-captives are gaining popularity, he said, because, “they are a way to take the pooling concept across state lines.”

Mr. Wightman explained that the difference between a rent-a-captive and your own captive is that a captive has to be capitalized “with your own money and/or letter of credit, and you are actually setting up an insurance company.”

A rent-a-captive takes advantage of the concept of a captive, “but your are renting capital and basically renting space,” he said. “So there are higher fixed costs, but the rent-a-captive does not require the upfront capitalization and the administrative set-up.”

Reproduced from National Underwriter Property & Casualty/Risk & Benefits Management Edition, September 16, 2002. Copyright 2002 by The National Underwriter Company in the serial publication. All rights reserved.Copyright in this article as an independent work may be held by the author.