The psychological evidence is quite clear: using a cell phone while you're driving is distracting. Conversing with someone on the phone imposes a cognitive strain that makes it harder for the brain to concentrate on other tasks. Hands-free systems keep drivers eyes on the road, but they don't do much to reduce their level of distraction. No matter how you cut it, the case for banning drivers from using mobile phones is a strong one.
What's less clear is whether or not these bans actually reduce road accidents. Given that mobile technology is fairly new, the question hasn't received much empirical attention. One study to look at texting bans in four states, back in 2010, actually found that accidents increased in those states, compared to neighboring states without the bans — perhaps because drivers tried to hide their phones while texting, making the act even more dangerous.
Economists Rahi Abouk and Scott Adams of the University of Wisconsin-Milwaukee recently examined the matter more broadly, conducting the first nationwide study of texting laws and fatalities. Their results, set for publication in next month's issue of the American Economic Journal: Applied Economics, are very instructive for public policy. Abouk and Adams report that texting laws only reduce fatalities under certain conditions — and that without strong enforcement, behavior relapses within just a few months.
Abouk and Adams gathered national data on road fatalities from 2007 to 2010 and paired it with laws from 49 states (Alaska's data was incomplete). They focused on one type of accident most likely to be caused by distracted driving: single-vehicle, single-occupancy crashes. They were aware they'd miss some multi-car crashes this way, but thought that if the bans do work, then these single-vehicle accident figures would be the ones most subject to change.
Next the researchers divided the texting bans by quality. Some states had "weak" bans: they only made texting illegal for certain subsets of the population, or made texting a secondary violation (meaning police couldn't stop a driver for that reason alone). Others had "strong" bans, which universally outlawed texting and made it a primary violation. Still other states, of course, had no ban at all. Abouk and Adams also considered whether states had concurrent bans against all handheld mobile usage.
They found that states with "strong" bans experienced an 8 percent reduction in fatal single-occupancy, single-vehicle accidents following the texting ban. States with "weak" bans, however, didn't show any significant improvement. But the effect didn't hold for long. Abouk and Adams report that fatalities rebounded in the months following the ban — and by the fourth month they had reached previous normal levels.
This rebound effect was sharpest in places without concurrent handheld bans. That suggested to Abouk and Adams that police have a hard time enforcing the texting law in the absence of a universal ban against handheld use. This conclusion makes sense: consider the difficulty of distinguishing a driver who's illegally texting and driving from someone who's legally driving and dialing. If drivers are allowed to use their phones sometimes, enforcing specific types of phone use becomes nearly impossible.
So, to recap, "strong" texting bans do reduce the type of accidents likely to be caused by distracted driving, but this public safety improvement diminishes quickly in the face of poor enforcement.
Several clear policy implications emerge here. The first is that making texting a "secondary" violation is about as good as not banning it at all. The next is that even states that make texting a "primary" offense must maintain heightened enforcement to sustain the benefits of the law. And the third is that the easiest way to facilitate strong enforcement is to ban handheld mobile usage in general. According to the Governors Highway Safety Association, only 9 states (including D.C.)* currently meet all these criteria. As more good research like this study makes its way into the public, one would hope to see that number climb.
* California, Connecticut, Delaware, D.C., Nevada, New Jersey, New York, Oregon, Washington. This list was limited to states with "strong" bans (e.g. primary offenses, universally applied) on both handheld usage and texting. West Virginia will join the list in July when its secondary handheld ban becomes primary.