Norm has been in an argument. He criticized John Goekler for asserting that anti-terrorist measures are a “complete waste”, on the grounds that there’s a distinction between dying accidentally and being murdered. Rob, however, rejects this:
If I had to die at 10:00 tomorrow it would be of no consolation to me that I would die in an accident rather than a terrorist bombing. Death is death.
To which Norm replies:
People take exception to being done in by violence, even if the event is a rare one. And because they do, they like to know that those who govern will take steps to protect them from this.
What all three are missing is the economist’s angle. This says that, for policy purposes, what matters isn’t so much between how people die, but how cheap it is to prevent deaths.
In principle, it is possible to completely eradicate some forms of death. We could make roads safe by banning cars. We could stop terrorists blowing up planes by banning air flight. We could stop husbands murdering wives by putting a policeman in every marital home for 24 hours a day.
But we don’t do any of this because it would be prohibitively expensive. We therefore allow some people to die because it’s too expensive to save them.
The relevant distinction, then, isn’t between terrorist deaths and accidents, but between deaths that are cheap to prevent, and deaths that are not. Insofar as Mr Goekler’s argument makes sense, it’s because of the view that, given the rarity of deaths from terrorism, the costs of stopping them exceed the benefits.
To make this judgment, we need to know two things. One is the production function of life-saving measures; how much, say, does airport security cost in terms of time, hassle and money and how many lives does it save? The other is the value of a statistical life; what‘s life worth?
On this point, though, a strange fact emerges. This value ranges enormously. W. Kip Viscusi has estimated that whilst some US regulations save lives cheaply - such as a ban on unvented heaters - others, such as limiting exposure to formaldehyde - cost billions of dollars to save a single life*.
And Australian research (pdf) shows that the value of a life is much higher for occupational health than for transport. We ban dangerous practices in the workplace even if it is very costly to do so, whilst allowing similarly dangerous practices on the road, even though it would be much cheaper to ban them.
In practice, then, it seems that policy-makers do draw the distinction Norm draws; they pay more to stop deaths in some contexts than others.
What troubles me, however, is that this might only be a happy accident. I fear that squeamishness about explicitly applying cost-benefit analysis to saving lives can lead to both wasteful regulation and needless deaths.
* Because they save so few lives even over many years, rather than because they cripple the economy.
In principle, it is possible to completely eradicate some forms of death. We could make roads safe by banning cars. We could stop terrorists blowing up planes by banning air flight. We could stop husbands murdering wives by putting a policeman in every marital home for 24 hours a day.
But we don’t do any of this because it would be prohibitively expensive. We therefore allow some people to die because it’s too expensive to save them.
The relevant distinction, then, isn’t between terrorist deaths and accidents, but between deaths that are cheap to prevent, and deaths that are not. Insofar as Mr Goekler’s argument makes sense, it’s because of the view that, given the rarity of deaths from terrorism, the costs of stopping them exceed the benefits.
To make this judgment, we need to know two things. One is the production function of life-saving measures; how much, say, does airport security cost in terms of time, hassle and money and how many lives does it save? The other is the value of a statistical life; what‘s life worth?
On this point, though, a strange fact emerges. This value ranges enormously. W. Kip Viscusi has estimated that whilst some US regulations save lives cheaply - such as a ban on unvented heaters - others, such as limiting exposure to formaldehyde - cost billions of dollars to save a single life*.
And Australian research (pdf) shows that the value of a life is much higher for occupational health than for transport. We ban dangerous practices in the workplace even if it is very costly to do so, whilst allowing similarly dangerous practices on the road, even though it would be much cheaper to ban them.
In practice, then, it seems that policy-makers do draw the distinction Norm draws; they pay more to stop deaths in some contexts than others.
What troubles me, however, is that this might only be a happy accident. I fear that squeamishness about explicitly applying cost-benefit analysis to saving lives can lead to both wasteful regulation and needless deaths.
* Because they save so few lives even over many years, rather than because they cripple the economy.
Comments