16 November 2016

Drunk Driving Laws and Disproportionate Punishment

Stephen Kershnar
Are Drunk Driving Laws Too Harsh?
Dunkirk-Fredonia Observer
November 15, 2016

Drunk driving occurs when someone drives a motor vehicle while under the influence of alcohol or drugs. If a driver is over 21, it is illegal for him to drive with a blood alcohol content (BAC) of .08 (g/dl) or greater.

Drunk driving kills people. In 2014, 31% of the U.S. traffic fatalities for the year (roughly 10,000 deaths) involved a drunk driver (BAC of at least .08). Most of those who died were the drunk drivers (64%).

The bulk of these accidents involved very drunk drivers, not buzzed ones. Roughly seven out of ten of these accidents had a driver with a sky high BAC level of .15 or higher. This is relevant as the .08 limit is easily met. According to the National Highway Traffic Safety Administration, a light woman is over the limit if she drinks two beers in an hour and a medium sized woman hits it with three. In 2013, the National Transportation Safety Board recommended the limit be reduced still further to .05. A medium sized male would be over the limit if he drank two beers in an hour. 

This low threshold results in a large number of people being arrested for drunk driving. In 2014, over 1.1 million drivers were arrested for driving under the influence. That is roughly one arrest for every two hundred licensed drivers.  

The penalties for drunk driving are harsh. In New York, the penalty for a first-offense DWI, where the driver has a BAC of .08% to .18%, results in an administrative license suspension of at least six months. The suspension is one year for drivers with a BAC of .18% or more or for drivers under 21 years old. The overall cost of such a conviction is probably around $10,000. This includes lost wages, attorney’s fees, court-ordered fines, car insurance increase, traffic school, substance abuse education courses, Department of Motor Vehicles fees, ignition interlock devices, towing and storage, and bail. Other states have more severe punishments.

These punishments are too harsh. In general, a punishment should be proportional to the crime. More specifically, punishment should be proportional to the offender’s blameworthiness or, perhaps, the product of his blameworthiness and the harm to another that he caused or risked causing. 

Per incident, driving drunk is not dangerous. In 2009, using data from the U.S. Census Bureau, there were 1.1 traffic deaths per 100 million miles driven. According to  Christopher Ingraham writing in The Washington Post, when legally drunk (0.08% blood alcohol content), one’s chance of being in an accident involving a traffic death triples, but even then, 3.3 traffic deaths per 100 million miles driven is very small. To put it in perspective, the average licensed driver driving drunk whenever he drives would, roughly, have to drive for more than 2,000 years before being involved in a traffic death. Another way to put it is that a person who drives drunk every time he drives and who drives the average distance per driver per year would have to go more than 27 lifetimes before he was involved in a traffic death.

            Here is another way to see the danger using older numbers. In Confronting Drunk Driving, H. Lawrence Ross points out that the chance of dying in a five-mile car trip is about one in ten million. The typical driver with a blood alcohol level of 0.1% is five times more likely than someone sober to be in an accident. Since the initial risk is so incredibly low, anything that multiplies that risk even tenfold would not create a substantial risk. Note that this likely overstates the risk as the fatalities per mile driving are decreasing. According to an article by Ken Bogen in Risk Analysis, the increased risk of drunk driving relative to sober driving is equivalent to that imposed by a nighttime drive by a seventy-year-old as compared with a daytime drive by a forty-year-old.

Not only is drunk driving not dangerous per incident, it is common. 20% of drivers self-report having driven after drinking and 8% of drivers self-report having driven drunk.

            Consider whether people who drive moderately drunk are blameworthy. Among the large number of drunk drivers, many people don’t consider the risk or judge the risk worthwhile and thus lack blameworthy mental states. Even if we focus on negligence, most people do not think that driving tired or night-driving for three hours or more is unjustifiably dangerous and yet these pose roughly the same risk as driving drunk.

Here is another way to see this. We generally don’t think the state should severely punish people for driving tired or night-driving for three hours or more. In most cases, drunk driving (when the driver is not very drunk) is similar to driving tired or at night for three hours. Hence, in most cases, the state should not severely punish drunk drivers. Similarly, in a 2006 study in Human Factors, David Strayer et al. found that using a cell phone (talking, hitting buttons, or texting) is as dangerous as drunk driving. Yet the penalties for the latter are much lower. This is even true for texting while driving.

An objection is that drunk-driving harm can be efficiently deterred through a very severe punishment that is infrequently applied. If a harm to another can be efficiency deterred through very severe punishment that is infrequently applied, then the state should impose such a punishments. Even if the punishment is efficient, it might still be disproportionate. The state can likely cut down cheating on taxes, hiring illegal aliens, and car theft by instituting mandatory twenty-year sentences for these crimes, but this penalty would be too harsh. Also, the life-saving benefit of severely punishing drunk driving has to be weighed against its cost in the same way that the benefit of a life-saving 55 mph speed limit has to be weighed against its cost.


The drunk driving laws are too harsh and disproportionate to the crime. 

No comments: