The Older Paper: ...And A Happy New Year!

My father, always eager to shape my young mind, used to joke that "30% of car crashes are caused by drunk drivers - can't we get all those dangerous sober drivers off the streets?" That kind of thinking is known today as the base rate fallacy, but apparently that problem's already been described by Plato. Or maybe it was Aristotle.

But quantifying how much more dangerous drunk drivers actually are is rather tricky. The best solution that I know of is a paper by Steven Levitt and Jack Porter. Here's the abstract:

We present a methodology for measuring the risks posed by drinking drivers that relies solely on readily available data on fatal crashes. The key to our identification strategy is a hidden richness inherent in two car crashes. Drivers with alcohol in their blood are seven times more likely to cause a fatal crash; legally drunk drivers pose a risk 13 times greater than sober drivers. The externality per mile driven by a drunk driver is at least 30 cents. At current enforcement rates the punishment per arrest for drunk driving that internalizes this externality would be equivalent to a fine of $8,000.

To apply their estimation technique, they have to make some assumptions which can be questioned, but, to their credit, they include estimates based on alternative assumptions. Recommended reading for all stats nerds.

May you have a funky new year's eve. And not make any stupid decisions ;-)