Finite Challenges Prompt New 'Insurance' Definition

I searched back through the document to check it out. Sure enough, there it was on page 17, along with examples of policies that are "unequivocally insurance"–homeowners and personal auto.

Ha! Does Consumer Reports know about this?

Now, I know full well there are many logical reasons for such a definition, but instead of trying to come up with serious reasons, let's indulge my silliness for another minute.

If insurers write contracts that are unlikely to result in any claims, shouldn't the premiums be a lot lower? Insurance rates are basically the sum of expected loss costs, expenses and a margin for profit. If no claims are expected, isn't the expected loss cost zero?

Okay, let's move to expenses–commissions. Oh, never mind. We better lower those if we're going to get rid of contingents. So the policyholder then is being asked to pay what's left–a few administrative expenses and a profit load–and insurers deserve to make a profit, after all, for agreeing to, uh, pay claims?

No, that's not quite right. Policyholders pay for the promise that insurers will pay claims, yet for the purposes of this technical definition, carriers actually don't think they're likely to pay claims if they're writing policies that are "unequivocally insurance."

As a consumer, I don't think I should cough up my hard-earned dollars for such an empty promise, unless compelled to by some state law. And if any lawmaker thought it reasonable to pass a rule forcing consumers to pay insurers to write policies that they didn't expect to ever pay anything on, we'd vote those folks right out of office.

That's enough silliness. Even the most ardent consumer advocates will find flaws in my tortured logic.

More seriously and philosophically, one might argue that an underwriter writing a single policy does not expect to pay a claim on that individual policy–or he or she wouldn't write it. But, of course, a collection of policies produces an average level of expected claims that is figured into the premium calculation.

The ITC explains the "not-likely-to-result-in-claims" part of its definition parenthetically, noting that "for life insurance, although death is certain, the timing of the death and the existence of insurance coverage at the time of death are not."

With the ITC definition, I believe FASB is trying to make a distinction between contracts that should be treated more like loans–those for which expected payouts or cash flows can be reasonably determined and therefore funded–and those for which the timing and amount of cash flows is uncertain. Within the latter group of "insurance" contracts are those "unlikely to result in any claims," in the ITC's language.

I'm not in any way trying to diminish the hard work that went into putting together the ITC document. Still, I can't help but remember a time when the definition of insurance was simple–with references to risk sharing and fortuitous events.

Why has all this work been heaped on accountants? What has brought them to the point of having to redefine insurance? It's the simple fact that a few people forgot what is "unequivocally not insurance."

It is not insurance when a company issues a policy that obligates it to pay claims, and then enters into a side agreement that erases the obligation. That is what brought the whole problem of accounting for finite insurance and reinsurance contracts to the offices of regulators, as Mr. Licht pointed out at the Casualty Actuarial Society's Seminar on Reinsurance in June.

He noted that all of the restatements related to finite reinsurance that have been announced to date "came about because there was no real intent to transfer risk."

"While there may have been a contract, there was also other documentary evidence–e-mails, even letters sometimes, faxes, notes written by underwriters in files–that indicated that the reinsurer was going to be made whole in one way or another," he said, adding oral side agreements to the mix. In some cases, he said, the ceding company would promise to buy additional coverage or pay additional premium, or they might commute contracts at noneconomic terms.

That's how accountants moved away from struggling to try to determine if it is reasonably possible (the chances are more than remote) that an insurer might incur a significant loss–the old definition of risk transfer. Now they are focused on what would seem to be the polar opposite–to define insurance as a contract under which you're unlikely to pay claims.

To add to my confusion, I started to think about finite reinsurance contracts like adverse reserve development covers, environmental cleanup cost caps and even those "loss mitigation" devices to cover overruns on litigation already in progress.

The writers of those policies certainly believe their contracts are unlikely to result in claims, don't they? They expect reserves to be sufficient, cleanup costs to be accurate and victories on lawsuits. So aren't finite contracts "unequivocally insurance?"

Sometimes, it's better to laugh at these things than to try to puzzle them out.