THE CATASTROPHE PROBLEM.(The Talk of the Town)

Akash Kapur reports from the coast of South India

From 1951, Rachel L. Carson on the sea

How to help

Markets are heartless. And so, last week, as the world was dealing with the human consequences of the tsunami that devastated South Asia, investors were pondering the financial consequences--especially for insurance companies. But the pondering did not go on for long. It quickly became clear that, despite the inconceivable number of dead and the obliteration of entire coastlines, in dollar terms the tsunami's impact would be relatively small.

The obvious reason is that the affected countries are relatively poor, which means that the people and the businesses there are less likely to have insurance. But even wealthy nations are not well insured against disasters such as this one. The earthquake that hit Kobe, Japan, in 1995 did more than a hundred billion dollars in damage, yet, although the city sits atop a known fault line, only three per cent of the property in the area was insured against earthquakes. Even before the tsunami, this past year--in which there were four big hurricanes in Florida, a calamitous Asian typhoon, and another Japanese earthquake--was the most expensive in history for insurers. They paid out forty-two billion dollars in disaster claims, and that was less than half the total damage sustained. In recent years, the cost of catastrophes has risen astronomically, but our ability to insure ourselves against them has hardly improved.

Catastrophes differ from the kinds of things we usually buy insurance for, and not only because they affect a lot of people. Insurance, at its core, is an actuarial business. It's about distributing risk over a large group of people, whose behavior is collectively predictable. Insurers can't predict what will happen to one person, but they know (or should know) what will happen to people in general--a certain number will get cancer, a certain number will die in head-on collisions, a certain number will live to be ninety-three. The business, grim as it is, is relatively certain, as long as you get the math right.

Catastrophes, however, are not nearly as predictable. They occur so rarely that it is difficult to discern patterns or make reliable forecasts. It's not that all natural disasters are unforeseeable--scientists have got better at forecasting hurricanes, and geologists can speculate about the likelihood of earthquakes over a long stretch of time--but there is nothing actuarial in these calculations. (The same goes for man-made catastrophes, like September 11th or Bhopal.) And the amount of damage inflicted by such disasters is even harder to predict. A tiny variation in a hurricane's path through Florida can be the difference between a few billion dollars in damage and a hundred billion dollars.