The Problem With "Taxing" Drunk Driving

Nate Silver has a good statistician's idea. What is a "statistician's idea"? I just made that term up. But it means an idea that, by the numbers and when plugged into any fancy model or regression analysis (you can tell I only have the vaguest notion what these things are), makes real good sense. While I'm no big city mathematician, but just a simple, country cognoscente, I have a problem with both the original idea and the new and improved idea. The original idea was to tax liquor a bit more and use the proceeds to pay for national healthcare. My problem with that is Silver's, too: drinking, as such, isn't really a societal ill. What is an ill is being drunk and, in addition (whenever you get an if-then, it's hell on applicability), being out in public and operating a vehicle or picking fights or, if in private, having physical altercations with family members, and such. Besides, how is drinking to be singled out as the first, best source of healthcare revenue? How about taxing fatty foods, or taxing people who don't exercise or whose BMI's are in the "Whoa, fatty!" range? (BMI, though, is total horse shit; it was invented by a 19th century mathematician, Adolphe Quetelet -- actually born in 1796! -- using body data from scrawny immigrants and laborers. So yeah, it accurately describes the Belgian ditch-diggers of 1835. And?)

But the proposed remedy, to tax drunk drivers $8000, is hardly better in practical terms. Silver acknowledges this, but on the grounds that people will stop getting DUI's if it costs $8000. The very beginning of this argument is flawed, however, because Silver bases his math on the number of arrests for DUI, not the number of convictions. As anyone who's ever been or known someone who has been arrested for DUI, or anyone who's walked around a college town can tell you, people fight DUI charges pretty hard. There are lots more lawyers out there getting DUI charges dropped than doing just about anything, judging by the number of shingles in Bloomington, Chapel Hill, Urbana, New Haven, and so on. Moreover, DA's often throw the cases out and, even when one goes to trial, the defendant often escapes with a suspended license and a relatively paltry fine -- less than 10% of $8000 anyway. DUI, second to assault, is becoming our national tragic joke if it isn't already.

Now imagine how much harder people would fight a DUI charge if it cost $8000. Imagine how loathe DA's would be to prosecute those cases if they knew that the defendant would spend up to, in theory, $7999 to get out of the charge.

Like I said, this is a very good statistician's idea. But it doesn't seem to have a lot of practical merit.