Earthquake

If you have never had the experience of being on a massive conference call while an earthquake is occurring at one of the call venues, I heartily recommend the experience.

Listening to an east-coast person talk through a lot of background noise from the folks in a tower in Los Angeles as they rode out a 5.8 ’quake was oddly amusing.

By now you’ve probably heard plenty about the 7.9 earthquake in the Sichuan Province of China — over 13,000 people dead, schools collapsed, chemical leaks…and it was felt over 1000 miles away.

The New York Times has a map available to provide a bit of scale around the potentially devastated area.

Looking at the map, and some of the descriptions, I can’t help but think — in terms of magnitude and geographic scope, doesn’t this sound a little like the New Madrid quakes of 1811-1812?

True, in terms of human tragedy, there’s no comparison — the central part of North America was very sparsely populated at the time, while over 15.4 million folks felt the Dujiangyan quake “strongly”.  But still… folks in St. Louis and Memphis…and property insurance underwriters and actuaries…ought to take a moment to contemplate the “what if’s”.

Seen at Insurance Journal:

In a special report A.M. Best Co. warns that “lurking beneath Central Europe are numerous faults that rarely break the ground surface but that still could lead to large scale catastrophe losses for insurers and reinsurers.”[...]

“A quake in Central Europe would not have to be as powerful as the 1906 San Francisco earthquake or the 6.7 magnitude Northridge, California quake in 1994 to generate at least $10 billion in total losses,” said Best. “This is due primarily to the growing concentration of property values exposed, the quality of existing building stock and the tendency of the soil in some regions to liquefy.”

Despite the growing risks, Best found that “penetration rates for earthquake insurance remain low. In countries such as Italy and Turkey, less than 30 percent of the total expected losses from earthquakes likely to happen once every 200 years would be insured.”

I wonder if saying “at least we don’t have any European quake exposure” would be a useful line to bring up when renegotiating reinsurance contracts.  (Probably not…)

Seen in Insurance Journal:

Figures from the state Department of Insurance show that earthquake coverage was carried on fewer than 38 percent of the insurance policies for homes, mobile homes and farms last year.

That was down more than 5 percentage points from 2001. During that same time, the average cost of residential earthquake coverage in Missouri rose by 15 percent, according to the department’s figures.

In response, Gov. Matt Blunt created a task force to study the availability and affordability of earthquake insurance, especially near the New Madrid fault.

RiskProf offers up a theory for the cause—”the expectation of a Katrina-type-federal-bail-out (KTFBO)”.

The effect of potential KTFBOs (cat-fee-bows) on insurance markets will be difficult to measure and I am not suggesting that this is the only reason for the drop in Missouri earthquake coverage, but it should not go unexamined as a possible reason. If KTFBOs are real, there is nothing the state government can do short of mandatory coverage and we know how popular that option would be.

I can buy that as a contributing factor. I can also offer a few other theories:

• The article cites a 15% increase in quake rates, which RiskProf observes is in line with the rate of inflation. However, it’s worth noting that that’s a statewide average, and it leaves open the possibility of some rate shock in southeastern Missouri, as some insurers started looking at risk aggregation and 500- or 1000-year PML’s.
   
• The exposure to quake risk isn’t uniform across the state. The Office of Social and Economic Analysis at U. Missouri has some pretty maps documenting by-county population changes between 2000 and 2005. If you eyeball the charts and do a little hand-waving…I could see that less-quake prone parts of the state might be growing faster than counties closer to the New Madrid fault zone. If you figure that quake take-up rates are less in, say, St. Joe or KC than they are in New Madrid and Carruthersville…this could easily be part of the explanation.
   
• However, the possibility that really struck me is wondering if this is an artifact of the real estate boom. Granted, I’m not that familiar with Missouri’s real estate markets, but it’s not too difficult for me to imagine that in the first part of this decade, folks were more tempted to buy houses at the limit of affordability. Among those consumers, I could easily imagine decisions being made to shave off seemingly “unnecessary” coverage, to save a few bucks and perhaps to get the ratio of monthly housing expenses under a certain ratio.

If the Governor wishes to see EQ coverage take-up rates rise, perhaps he should commission some assistance from Doctor Iben Browning, whose work convinced me to take a day off from studying and go walk along a levee near New Madrid, Missouri on 2 December 1990.