This week’s earthquake in Christchurch could be the worst insurance catastrophe since Hurricane Ike hit Texas in 2008, according to early reports.
AIR estimates insured losses between US$3.5B and US$8B. JP Morgan estimates $12B. (Ike came in at $12.6B in the U.S. alone, $20B including all Caribbean basin countries, with all numbers updated for inflation).
I don’t have access to the letter to clients that contains Morgan’s estimate, but it sounds a bit like the analyst doubled the loss estimates from September’s New Zealand earthquake, which looks like a $5B to $6B event.
AIR’s original estimate for the September earthquake was $2B to $4.5B, and reinsurers spent much of the fourth quarter raising their estimates on that loss. Bumping AIR’s $4.5B up a third gives you $6B. Doing the same to the $8B estimate gets you around $10.5B.
Before being too harsh on the cat modelers, I should note Scientific American reports the fault whose shifting caused both earthquakes was unknown before it, well, started shifting last year. (h/t Risk Market News via twitter)
I should also point out that AIR’s estimates exclude demand surge – when construction costs soar because raw materials and contractors increase prices because labor and material are in short supply. And I can’t think where demand surge would be greater than an antipodean city of 400,000 struck by two major earthquakes in six months.
So there’s a case to be made that the $12B estimate is low. I think it’s fair to say the city would take decades to rebuild without the reinsurance industry.
For some perspective, $12B is equal to around 10% of New Zealand’s GDP. That would be equivalent to a $1.4 trillion insured loss in the United States.
How bad is that? An asteroid striking Manhattan could create $1.2 trillion in property losses, RMS has estimated (pdf), though that number appears to be total property losses. Insured losses would be somewhat less.