ASK A CASUALTY ACTUARY

What are the most common pitfalls to avoid in comparing risk financing program options?

* Unfortunately, there are several common pitfalls. They tend to fall into two categories because there are two key areas that should be assessed in comparing options for coverage or risk financing.

Those areas are:

* A comparison of expected total costs.

* An assessment of the degree of risk posed by different options.

Let's look first at common pitfalls in making comparisons of expected costs. Such costs often are estimated by taking recent losses and adjusting them to the current year or the next future year. Properly doing this usually requires making numerous adjustments. Among them are those needed to reflect the effects of claims inflation from the year the claim was incurred -- that is, the year the accident occurred that gave rise to a claim -- to the prospective year. Similar adjustments typically are needed to reflect changes in the number of exposure units -- for example, payroll, sales, number of vehicles, budget, etc. -- and in claim frequency trends. Further adjustments are needed to reflect loss development and the appearance of late-reported claims.

Incurred amounts from claims tabulations are only the starting point for deriving unbiased estimates of ultimate losses and other program costs.

Here is a classic example: Should an excess policy be purchased for claims exceeding $500,000? A roster of all past claims of more than $100,000 is reviewed.

The largest claim incurred in recent years is only $330,000. Therefore, it is concluded that very little should be paid for such an excess policy.

However, if all of these past claims had been adjusted for inflation in the size of claims, then the trended size of four claims would exceed $500,000.

Furthermore, if the fact that the number of exposures had grown very dramatically is reflected, those four claims should really be viewed as representing nine prospective claims that would be covered by the excess policy. This changes the whole tenor of the total cost comparison.

Another common pitfall is assuming that your own large claims experience is a reliable predictor of future excess experience. The main problem is often that there are so few large claims that whatever indication you derive from them is subject to tremendous variation on the basis of the chance occurrence or absence of any one or more of those few claims. From a statistical viewpoint, what has been derived is unreliable and basically meaningless. To get around this problem often requires considering information from a large number of risks similar to your own regarding the relative proportion of large vs. small claims -- or of losses in some excess layer vs. losses in some primary layer. Another way of tackling it is to fit a mathematical distribution to the number of small-and medium-size claims to derive an expectation of the number of large claims.

Another pitfall is failing to reflect shifts in the types of losses that have been occurring. New kinds of causes of loss may be becoming more frequent in current times -- or the reverse could be true.

It may also be important to reflect changes in what different policies might cover in making cost comparisons. Key exclusions or endorsements can have a dramatic effect on what will or will not be covered by a prospective policy.

It should go without saying that making a fair comparison of alternatives normally requires reflection of the time value of money. This is most commonly done by estimating the present value of the total costs of each alternative.

Another entire area of common pitfalls involves the matter of reflecting the relative degree of risk that various options pose. It is not uncommon for this factor to not receive any explicit or implicit recognition. All that is done is a comparison of total costs on an expected -- or best estimate -- basis. This ignores the substantive value that risk transfer can have in dramatically reducing the degree of risk you are facing.

The issue of whether or not to transfer risk is really a question of weighing the risks of self-insuring against the rewards -- expected cost savings -- of self-insuring. The risk is that your total costs could significantly exceed the total costs of standard insurance. In today's soft market, it is not uncommon for the total costs of guaranteed cost coverage to be less than the expected costs of self-insurance. When that happens, the decision is a no-brainer. What is the point of assuming more risk if it is going to cost you more to do so than giving that risk to an insurer?

Of course, there are many extenuating circumstances that may still make self-insurance the better option even when insurance premiums are unusually attractive. One key consideration is the potential impact that a change in who adjusts claims may have on total costs.

When cost comparisons are made only at the expected level, then the consequences of a wide range of alternative scenarios are ignored. It can be very important to evaluate how different options compare in the event that total losses are such as would occur during a worse-than-average year or a catastrophic year.

It can be tempting to boil down the issue of risk to just one number. It is rarely as simple as that.

Let's look at another example. Suppose that the projected cost of self-insurance is $8.3 million, as compared with a guaranteed cost premium of $6.8 million. This comparison leaves the strong impression that continued full insurance is the obvious choice. However, the $8.3 million projection includes a $1.6 million "margin for adverse experience," which is based on the difference between the 85% confidence level estimate and the expected estimate.

If we accept this margin for adverse experience as our best assessment of the degree of risk presented by self-insurance -- note that there are many other measures -- then you are risking $8.3 million minus $6.8 million, or $1.5 million, by self-insuring. The potential reward for self-insuring is the positive difference, if any, between the expected total costs of self-insurance and full insurance. Suppose that difference is $6.8 million vs. $6.5 million, or $300,000. This means that you would have been risking a possible $1.5 million loss in order to earn an expected return of $300,000.

That works out to a 20% rate of return. Because this is a quite attractive rate of return, self-insuring would probably be the best choice -- all other things being equal. This is opposite to the recommendation suggested by the first approach.

Assessing the relative degree of risk posed by various alternatives does not have to be an entirely subjective process. There are several ways of measuring the degree of risk of each option and of systematically assessing the value of risk transfer.

Would you like advice from an experienced colleague on a risk management, benefits management or actuarial problem? Four quarterly features in the Perspective section of Business Insurance can give you some answers.

Ask A Benefit Manager, Ask A Risk Manager, Ask A Benefit Actuary and Ask A Casualty Actuary answer written questions from readers on risk and benefits management issues and actuarial problems.

This month's column on actuarial issues in the casualty field is written by Richard E. Sherman, president of Richard E. Sherman & Associates Inc. in Ashland, Ore. Dennis J. Nirtaut, managing director of compensation and benefits for Arthur Andersen & Co. L.L.P. in Chicago, answers questions for benefit managers. Christopher E. Mandel, senior director of risk management at Tricon Global Restaurants Inc. in Louisville, Ky., answers questions on risk management issues. William J. Miner, an actuary with Watson Wyatt Worldwide in Chicago, answers actuarial questions on benefits issues.

Address your questions to ASK, Business Insurance, 740 N. Rush St., Chicago, Ill. 60611. Please give us your name, title and employer; however, Business Insurance will consider unsigned letters.