With a handful of insurers revising their individual company loss estimates in recent months, none of the sources contacted by NU saw any immediate need to go back to the drawing board to make wholesale revisions to the current industrywide estimates.

“In the greater scheme of things, were probably talking about nickels and dimes relative to the grand total,” said Robert Hartwig, senior vice president and chief economist for the Insurance Information Institute in New York.

The “nickels and dimes” of recent months include: a $500 million boost in Munich Res estimate; $200 million for XL Capital; $100 million at Royal & SunAlliance; and AXAs $85 million.

“Theres always going to be some uncertainty,” Mr. Hartwig said. But since October 2001, company revisions have been sporadic and, he points out, “they have not been enormous relative to the total estimated loss figures out there.”

“This is nowhere near the sort of upward development we saw with [a Hurricane] Andrew or with a Northridge [earthquake],” he said. He noted, for example, that initial estimates for Andrew put forth by the Property Claims Services Unit of the Insurance Services Office were in the area of $4 billion–a figure that “quadrupled over a period of about four-and-a-half months.”

“With Northridge, certain types of damage didnt emerge for months. You dont uncover a cracked foundation in a house for years,” he said, noting that some Northridge claims have reopened. “But with Sept. 11, it was obvious what had happened. The buildings were destroyed. People were dead.”

He also said damage related to the Sept. 11 attacks was contained in a compact area”a few dozen blocks of lower Manhattan, not hundreds of square miles of South Florida or California.”

He noted that there were 700,000 claim files for Andrew, while property and business interruption claims for Sept. 11 only total 51,000, according to the latest report from PCS released in June. “When you add in all the other types of claims, theres potentially 70-to-75,000 claims (including life)–10 percent of what we saw in Andrew,” he said.

Unlike the other events, the history of World Trade Center industrywide loss estimates, so far, includes at least one significant downward revision–eliminating the high side of an early range from a September 2001 Milliman U.K. report. The only miracle to arise from the tragedy–a higher than anticipated number of survivors–explains the move by most analysts to discard that $70-to-$100 billion estimate.

David Sanders, Millimans U.K.-based consulting actuary, who was one of the authors of the report, recently confirmed that the firms initial estimates were made the day after the tragedy, at the Reinsurance Rendez-Vous de Monte Carlo, when early speculation put the death toll at nearly 10,000.

That assumption figured prominently in its estimates for liability and life insurance losses, with liability losses assumed to be $2 million per death, and life coverage roughly $350,000 per death, with some upward adjustments for key-man policies. Assuming 3,000 deaths (based on a figure provided by Mr. Hartwig, who calculates 3,056 deaths for the WTC and Pentagon attacks), Millimans overall estimate lowers to roughly $52 billion.

Mr. Sanders confirmed the appropriateness of such an adjustment in an e-mail correspondence with NU. “Another key factor in the current position is the speed of the cleanup. This was originally estimated by engineers [to take] 10-to-15 months, depending on New York weather. A mild winter has helped the situation,” he said, indicating that Milliman would “likely come up with lower estimates for the property damage and business interruption category, if it repeated the exercise today.”

Tillinghast-Towers Perrin, the other major consulting firm to put out a range of estimates in September–of $30-to-$58 billion–has not revisited its numbers, according to its representative.

Other experts, such as Mr. Hartwig and John Kollar, vice president of consulting and research for Jersey City, N.J.-based ISO, have analyzed estimates of the consulting firms, as well as those of investment analysts, adding their own assumptions to fine-tune the numbers. Mr. Hartwigs estimate is now $40.2 billion, and Mr. Kollar said ISOs range is $30-to-$50 billion.

Helping to solidify and tighten the ranges, “we now have a pretty good handle on what the property and business interruption exposures are going to be,” Mr. Hartwig said, referring to a $20.3 billion estimate put forth by ISOs PCS.

Mr. Kollar explained that the estimate is based on a survey approach. The firm conducts surveys of insurers, agents, adjusters and others, and then analyzes and trends the data gathered.

While a June re-survey prompted an increase from an earlier figure of $16.6 billion released in March, and although some of the most recent company revisions have been in the business interruption line, experts agree that liability loss estimates are now the softest numbers.

“The liability classes may take 10-to-15 years to settle,” Mr. Sanders pointed out. “Lockerbie has only just been finalized,” he said, referring to the 1988 terrorism air disaster in Scotland. “I suspect this [Sept. 11] will take as long as plaintiffs search for insurance cover from parties other than aviation insurers and airlines.”

Mr. Hartwig said that todays estimates of liability losses arent much different than they were last September, when Tillinghast put forth a $5-to-$20 billion range, calling the figures “essentially an educated guess.” He said current industry ultimate-loss estimates assume “a certain amount of litigation will arise,” as some victims and their families will opt not to accept compensation from the Federal Victims Compensation Fund. According to the Department of Justice's Fund Web site, approximately 150 victims representatives made Fund claims as of Aug. 27.

With so many others undecided, “How are you possibly going to set a liability number if you dont know the number of claimants?” asked Steven Visner, a partner in the Actuarial Services Group of Ernst & Young in New York. Beyond that, there are questions about whether property and workers compensation claims will ultimately turn into liability claims, with those insurers subrogating against “security companies for the airlines, Boeing, or whoever manufactures the cockpit door.”

“I dont know if they will, but they could,” he said, reporting that his firm, which is working on a study of subrogation activities of six major insurers, has raised the question. “They said they havent decided,” he reported.

While Mr. Visner conjectured that insurers might be reserving for liability exposures at maximum policy limits, Mr. Kollar took a different view. “As an insurer, you wouldnt know whether to put up a claim or not,” he said.

Turning to the general question of whether insurers have actually booked losses anywhere near the ultimate-loss estimates, Mr. Kollar said there seemed to be a reserve shortfall for U.S. companies at year-end. Basing his remarks on an ISO analysis of Annual Statement information, he said that U.S. insurers disclosed about $10 billion, on a net basis.

Assuming ultimate losses of $40 billion, and roughly half of that for U.S. companies, ISO believes there could be a $10 billion deficiency, he said.

NUs more recent compilation of insurer disclosures (see page 11) indicates a different split between U.S. and non-U.S. companies, putting the U.S. portion at about one-third, suggesting that only $3-to-$4 billion is unaccounted for.

“You may never actually get to that [ultimate] number in terms of reported losses,” said James Auden, senior director for Fitch in Chicago, who believes the ultimate loss figure is in the $30-to-$40 billion range. He referred to an October report by the rating agency that $3-to-$6 billion could disappear as a result of finite reinsurance coverages and drawdowns on equalization reserves by European reinsurers.

The report explained that primary insurers covered by finite contracts can claim recoverables in the full amount of ceded losses, but that offshore reinsurers can discount assumed losses, recognizing future investment income. The difference, then, will be unreported in industry totals.

Reproduced from National Underwriter Property & Casualty/Risk & Benefits Management Edition, September 9, 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.