Figures Don't Lie...

Earlier this year, local television news stations in my area repeatedly aired stories about a string of violent crimes in a park not far from my home. The stories about Flushing Meadows-Corona Park--probably better known as the site of the 1964 World's Fair--were broadcast with one particularly troubling image of a man who had been beaten beyond recognition.

As newscasters urged viewers to help identify him, they fueled local citizens' anxieties with news that there were 21 serious park crimes in the fourth quarter of 2006.

Countering these figures, the New York Police Department issued a statement, explaining that two men believed to be responsible for 11 crimes were arrested within a month. "Without them, park crime would have declined from the 16 in the previous (third) quarter to 10 in the fourth quarter," the NYPD statement said.

Continuing in this vein, of course, the NYPD could just as easily have pointed out that without a dozen criminals, park crime would have been nonexistent in the quarter.

This type of numerical slight of hand used to be all the rage in the property-casualty insurance industry. We all remember the typical phrases--"If not for adverse reserve development, record hurricane losses and a bad investment, we would have reported a profit."

Criminal No. 1--adverse development--is the one that captivated our attention for years with his ability to beat companies to a bloody pulp armed with asbestos liabilities, construction defect issues, excess workers' compensation claims and the like.

His alter ego--favorable development--has emerged more recently, as we note in the data analysis contained in this issue (see page 13). With his more benevolent spirit, he may not get all the press he deserves.

At a typical industry gathering early this year, for example, an industry spokesperson tried to convince investment analysts that last year's record earnings were sustainable. He pointed to the fact that insurers were managing their capital, suggesting this would stave off reckless soft pricing, and added that there was plenty of earnings' steam in unrealized investment gains.

But beyond a brief bit of lip service to healthy balance sheets, he never directly mentioned the possibility that there was some fat in carried loss reserves.

Sporadically, public companies started announcing fourth-quarter financial results at about the same time, and it soon became clear that some insurers were already finding redundancies in their loss reserve positions available to release into earnings.

Without performing rigorous actuarial analysis, it's not possible to know how much cushion exists for the industry, or how far insurers are from tapping them out.

Given the magnitude of profits from other factors--a lack of catastrophe losses being one of them--and given the amount of disclosure surrounding loss reserve estimates now being required of public companies, however, it should be safe to assume that real changes in development patterns prompted the initial takedowns, and not attempts to manipulate earnings.

While we hope prudent loss reserving practices continue, increases in the frequency of favorable development disclosures and expert opinions on overall reserve adequacy bear watching as harbingers of the next hard market.

No insurance executive will tip us off with a phrase like, "If not for favorable development, our reported 102 combined ratio would have been 105."

But history teaches that when premiums lag inflationary loss trends, underwriting profits fade. And without reserve cushions to hide the bad news, signals for needed price hikes will become clear even before the next mega-loss event.