Assessing Black Swan Events in the Insurance Industry

Insurance underwriters spend most of their time estimating risks and setting premium rates at a level that will result in acceptable underwriting profitability. One of the fascinating aspects of insurance is that premium rates may not always account for “black swan” events. Black Swan events refers to scenarios that are considered to be extreme outliers.

In the financial markets, many observers believe that the normal distribution does not accurately represent events as they are observed in the real world. For example, Benoit Mandelbrot’s book, The Misbehavior of Markets, goes into great detail regarding the failure of the financial markets to adhere to the normal distribution which forms the foundation of assumptions made in the efficient market hypothesis.

Unlike financial markets, insuring against natural phenomena may more closely resemble such a distribution. National Underwriter P&C magazine recently ran an article that reported on an estimate for damages that would result if a meteoroid were to hit Manhattan. If an asteroid explosion similar to the one that hit Siberia in 1908 were to occur above Manhattan today, damages would be extreme:

Risk Management Solutions (RMS) in Newark, Calif. for its estimate, superimposed the damage footprint of the 1908 event which saw a comet blast 3 to 6 miles in the upper atmosphere above the Tunguska River forest in Central Siberia.

That explosion leveled trees within a 770 square mile area. RMS said that such an event could create hurricane-force winds and fire ignitions due to thermal exposure.

For Manhattan and its surrounding area, RMS said total economic exposure would measure $760 billion between the outer and inner contours of the footprint and $1.38 trillion inside the inner contour. As a result, property losses are estimated at approximately $1.19 trillion.

The article links to a report with more information on the 1908 Siberia incident and detailed projections regarding the theoretical impact of such an event over New York City. While insuring against natural phenomena may lend itself to statistical models based on the normal distribution, the same cannot be said for manmade disasters. This is why insurers after the September 11 terrorist attacks have been forced to reassess their policies on insuring against terrorist attacks. In reality, only the Federal Government would be in any position to insure against a catastrophe of such magnitude whether it is caused by a natural event or by human evil.

Comments

[Ravi Nagarajan:]

With all due respect, you need to refine your marketing "strategy". Rule #1 in the minds of most people is to not insult your target audience.

On Nov 12 12:57 PM STOCKHITMEN wrote:
> Our clients made money, if you were not a donkey you could have the
> Black Jack too, instead you are the monkey reading this crap on SA.

Nov 12 01:27 PM

[TinyTim:]

Power law phenomona in nature, such as eathquakes and asteroids are more applicable to insurance underwriting than a normal distribution black swan.

The big question is how applicable the power law is to finance and economics.

Nov 12 03:43 PM