Earthquakes and the dismal science

As the 100th anniversary of the great San Francisco earthquake and fire approaches, I find my mind often dwelling on earthquakes. Berkeley is full of multi-story concrete apartment buildings built during the 1960s catering especially to students. These buildings are very top-heavy and without a seismic retrofit, are prone to collapse and "pancake" in the event of an earthquake with the floors compressing to within a few inches of each other. That is why it is predicted that a major earthquake in the Bay Area (there is greater than 50% chance this will happen within the next 30 years) will kill thousands of people.

So with these cheery thoughts in my head, I naturally start trying to apply the tools of the dismal science I am studying. Consider the following scenario: a building owner is liable for $1 million for every person that dies in his building if it collapses during an earthquake. Also, suppose that his building has 100 residents, 50 of whom would be expected to die in an earthquake. Roughly speaking, the probability of a major earthquake hitting in any given year is about 2% ((1-0.02)^30 is 0.55 so this is an underestimate). That means the expected payout in a given year is 0.02*50*$1 million = $1 million.

In other words, the building owner should be willing to pay up to $1 million annually for an insurance policy that protects him from liability. The punch-line of this is that often times the expense of such liability insurance is much cheaper than retrofitting the building to survive an earthquake.

What should public policy on this issue be? One option would be to force building owners to retrofit their buildings or else have the building demolished. Since the market for rental housing in the Bay Area is competitive and not all buildings are unsafe, owners of buildings not up to code might find it unprofitable to retrofit their buildings since they could not earn that money back by raising rents and instead allow the building to be demolished. Legally, the government would probably have to compensate building owners by the fair market value of the building as well.

Another possibility would be to increase the amount by which building owners are liable in the event of a building collapse. This could tip their decisions in favor of retrofitting instead of buying insurance. Again, if the rental market is fairly competitive and if insurance is mandatory, building owners would have to absorb most of the cost themselves and either abandon the building completely or retrofit without raising rents substantially.

Both of these options have costs associated with them and implicit in all of this is one of the most difficult issues in economics and moral philosophy: how much should we be willing to pay to prevent someone's death? Presumably, the number should be less than $10 trillion -- the value of U.S. GDP -- and certainly much greater than pocket change. The other topic implicit here is the notion that people in the real world ignore or belittle low-probability catastrophic scenarios. The fact that many residential, commercial and government buildings would become death-traps in the event of a 7.0+ magnitude earthquake here is public knowledge but the will to do something about it is quite small.