Terrorist attacks like those on Sept. 11, large-scale industrial accidents like Three Mile Island, hurricanes like Andrew, or earthquakes like the one in Northridge, Calif., that killed 60 people--these are all what economists call low probability, high consequence events. Making economic decisions about how to prepare for such "extreme events" is a difficult process. Under what circumstances are the benefits of strengthening a building against explosions or earthquakes worth the costs? A new study sponsored by the National Institute of Standards and Technology (NIST) offers strategies for finding answers to such questions.
Conducted by the University of Pennsylvania's Wharton School, the study* found that preparing for extreme events requires an understanding of risk "interdependencies." A security plan, for example, is only as strong as its weakest link. It also requires cooperation between public and private organizations because individuals and organizations often don't take actions to mitigate low probability risks unless there are incentives to do so.
Ultimately, the study authors concluded that dealing effectively with extreme events depends on a complex interplay between risk assessment, perception and management. Risk assessment for a power grid in Ohio needs to include possible negative effects from domino-like failures throughout the northeastern United States and Canada. People perceive risk more clearly when they understand its cumulative effects. More people will wear seatbelts, for instance, if told they have a 33 percent chance of an accident over a 50-year lifetime of driving than if they know there is 0.00001 percent chance for each trip. And risk management is more likely if the economics are attractive. A $1,500 loan to prevent flood damage is more affordable if payments are divided over the life of a 20-year mortgage and if insurance premiums drop as a result of the improvements.
* An electronic copy of Risk Analysis for Extreme Events: Economic Incentives for Reducing Future Losses by Howard Kunreuther, Robert Meyer and Christophe Van den Bulte is available at www.bfrl.nist.gov/oae/oae.html.
Cite This Page: