Comment by mikeash
6 years ago
People are bad at understanding risk. Tell someone that this change will save the company enough money to bump the share price by $2, but there's a one-in-a-million chance per flight that it will cause a crash that will destroy the company instead. Do you go for it? A lot of people will hear "one in a million" as "basically impossible" and say it sounds great. They'll probably rationalize it further by deciding that you're being paranoid about the risks. And then a while later you have a few hundred dead people and a company with a shattered reputation.
And to further complicate it, the smaller a risk, the worse we are at estimating it.
Investigations after the Challenger Disaster revealed that engineers at NASA thought the space shuttle had a 1/100 failure rate per mission, while managers thought the shuttle had a 1/10,000 failure rate.
From a completely technical perspective, one can understand the cost of this risk with real option valuation techniques. However I’d say this is an inappropriate method when so many lives are at stake.