Texting while Driving

Take this story with a grain of salt but it’s useful to keep complexities in mind when regulating:

The Insurance Institute for Highway Safety says that 3 of every 4 states that have enacted a ban on texting while driving have seen crashes actually go up rather than down.

It’s hard to pin down exactly why this is the case, but experts believe it is a result of people trying to avoid getting caught in states with stiff penalties. Folks trying to keep their phones out of view will often hold the phone much lower, below the wheel perhaps, in order to keep it out of view. That means the driver’s eyes are looking down and away from the road.

Very appropriate hat tip: Offsetting Behavior.


In the last 10 years as cell phones became universal and in the last 5 when texting emerged, have accident rates risen or fallen?

According to IIHS, insurance claims related to accidents have stayed the same since the emergence of cell phones. It could be that people will find ways to distract themselves while driving no matter what. If it isn't cell phones, it is eating, shaving or fiddling with the radio. Targeting one type of distraction appears to be ineffective either because it doesn't change behavior or because it channels behavior to other, equally dangerous distractions.


People who use cell phones while driving are less-safe drivers even when they aren't using their cell phones, according to research at MIT. I don't think this contradicts other work which says using your phone while driving is dangerous, but it does suggest there's a lot more going on, and maybe the cell phone effect is not as strong as we casually think.

In that case the optimal solution is to decriminalize cell phone use. Then a year later send out surveys asking if you've used a cell phone while driving more than 5 times in the past year. Take away the licenses of anyone who says yes.

Is it possible that people may be more dangerous driving with cell phones, but that they are able to drive fewer total miles, because the calls or texts stop unnecessary driving? Or that they slow down, so that even if distracted, others can better avoid them?

By itself this statistic should be irrelevant.

What if 4 out of every 4 states that didn't enact a ban have seen crashes go up? Where's the control?

Governments never run the test beforehand.

Or, to be precise they do. They just run their experiments on us without controls or any hope of repeal in the event of failure.

To Rahul, +1. Also, since the term is "increase", what is the level relative to other states and what was the rate of increase prior to enactment.

How 'bout we ask that before we pass the law that kills people?

Show me the study, Bill. The one where they looked closely before passing the law.

The "Don't be too hasty" comments get tiresome especially when it is "don't be too hasty to criticize some government action" that was itself clearly too hasty.

Andrew, In case you didn't notice, the Rahul's criticism was that the fact was IRRELEVANT. I also pointed out that other facts would make it IRRELEVANT as well.

Do you have some basis for arguing that the fact was relevant when there were no controls?

Exactly. Presumably, this law was enacted at a time when cell phone use and texting was increasing dramatically, as it has everywhere.

This. Accidents per text or some similar measure would be more telling.

I don't know, but I believe overall US vehicle fatalities, at least, have been consistently declining - absolutely, not just proportionally - for decades.

Raises another question: Was he measuring total-crashes or texting-related-crashes?

(BTW, you mention fatalities; he uses crashes. )

Tangentially related - I walk a lot for exercise and some of this takes me along major roads in Bethesda that feed the National Institutes of Health and the Navy/Walter Reed medical center. I get to observe drivers who are talking on cell phones and I have observed that the preponderance of those talking are women (perhaps 80%) and they are not using hand's free devices. I see about the same ration when I'm driving somewhere as well. This strikes me as an interesting number and of course it is from an n=1 observer. I wonder if others have seen the same thing.

Regarding cell phone use whether speaking or texting and accidents; it should be easy enough to check cell phone logs in the event of an accident and see by the time stamp whether the phone was in use or not.

Is the driving population you observe roughly 50-50 Male-Female?

The point that always intrigued me was the exemption for hands-free devices: Is the risk a dexterity issue or a cognitive issue? And I refer to cellphone-calling, not texting.

I recall a study that claimed hands-free didn't help.

The real issue, as Walter Block has pointed out, is that government roads are inherently unsafe. There will always be a market for the Walter Blocks of the world because the government possesses cognitive immunity in a lot of things. We blame the drivers. In engineering quality and safety you always blame the management.

Thus we will invent the self-driving cars as a workaround to avoid having to create a better road system.

"There will always be wingnut welfare for the Walter Blocks of the world because the Koch brothers possess a lot of things."


"the government possesses cognitive immunity"

If you were to ever take a position on the other side of this "government = moron" paradigm, I would be able to take you sirously. I mean just once, please!, as i find many of your arguments intriguing.

Yes, pretty much 50-50 mix in terms of gender.

I recall a similar study some time ago in Europe about speed cameras, with similar results. It is true that the control here is relevant, hence Alex's suggested grain of salt. The larger point however being that regulating something in of itself does not make it safe. 3 words: War On Drugs.

It is still legal in all states to drive while balancing a bowl of BOILING OIL on your knees. The futility of this sort of lawmaking is reaching levels of self-parody that continue to surprise me.

No one wants to do that. Why write a law about it?

More importantly where is the negative externality? Like with spikes sticking out of your steering wheel, this is likely to have positive externality. So why would you regulate it.

Gee, Robert, you've got me there...

I agree, we should regulate your intense desire to text while driving.

Perhaps these bans were put in place in states that were forecasting especially pronounced increases in handheld-device-related crashes in the coming years.

The short run elasticity may be different from the long run elasticity. Drunk driving laws probably did little to change people's behavior when they were first passed... but after many years we may reach a new equilibrium where people think texting and driving is as stupid as drinking and driving.

Drunk driving probably is the right analogy here; massively popular behavior, socially acceptable or even mandatory in many contexts, insanely dangerous if combined with driving.

What brought about a substantial decline in drunk driving was not laws saying "drunk driving is illegal". It was laws saying "drunk driving is illegal, and we actually will put your sorry ass in jail if we catch you at it", and judges following through by putting people in jail, and police enforcement activities specifically aimed at catching drunk drivers (e.g. checkpoints). And a parallel public-relations campaign to persuade people that no, this wasn't just the nanny state overreaching itself, this was a clear and present danger being addressed by the only effective means available.

We could certainly do the same thing with texting-while-driving. Probably easier, given that a cellphone intrinsically shouts "Hey, everybody, cellphone in active use!" at the top of its tiny electronic lungs whenever it is in fact being used. And if the IIHS speculation is correct, we'd probably want to trigger the initial stop on the signal rather than the highway patrolman's visual observation of a cellphone in use.

Whether we want to, depends on the scale of the danger we are dealing with. As has been noted, we could use better statistics on that matter,

Most (all?) cell phones have GPS. They could just not work above 10 mph, save maybe to dial 911. I understand that would hit passenger use as well - maybe there could be some technical way around that - but it seems preferable to jailing more people.

But, as you correctly state, the size of this problem is not really clear, and that should have considerable impact on what measures are seen as reasonable for solving it.

There's quite a lot of managers who employ drivers and travel as passengers precisely because they need to process their agenda while being on the way.

If cellphones did not work in moving vehicles, they would scream like hell - and justly so. I know quite a few such people, being occassionally one of them. Lost productivity would be significant.

Quite a lot of people have chauffeurs? I think this just isn't true. In any event, I said you would want technological solutions to allow passenger use.

I get the odd car service or taxi which is maybe close enough to count, so I'll bear with you. Perhaps there could be a port in the back seat that disabled the feature? Perhaps the phone could be disabled within a meter of the steering wheel? Perhaps there could be a tax, say $500 a year, for a >10 mph phone so the use was confined to people who could really justify it?

But anyway, it's not at all obvious that there's a benefit from the existing rules or the hypothesized new rules that justifies the costs. I only meant that it seems like there are technical solutions here, and they seem less invasive than the legal solutions that would really work, and more effective than the legal solutions which are purely symbolic.

That seems like an extreme measure for questionable benefit.

I think I agree, although the technological solution seems less extreme than throwing people in jail for it.

It's not obvious that if you could magically stop people from using cells while driving that there would be a benefit offsetting the lost utility of not being able to use your cell while driving. And in the real world, you can't use magic, your rules and enforcement impose costs beyond the forgone utility.

_If_ you could really understand that there's some negative externality from cell use while driving, you could just tax at-speed cell use and use the proceeds to offset the harms. Perhaps you could compensate victims, or help everybody afford bigger, safer cars. I can't believe the externality is as high as $5 per call. Yeah, you'd tax some passengers, too, but realistically, how common is that? Look around on your commute. You'll see numerous people talking and texting, but rarely will it be a passenger. Usually it will be the driver, as usually the driver is alone or the only adult in the car.

This is an important point. My impression, at least in New York, is that the laws about cell-driving and text-driving aren't being enforced. So I wouldn't expect them to reduce accident rates.

Ed, The law probably gets "enforced" in a way you are not anticipating: if someone is injured, there is a presumption that the violation of the law was the negligent act; presumptions matter in burden of proof situations, making it easier to sue the person who was texting. I don't know if NY is a negligence per se state, rather than a presumption state.

NY is a no-fault state, so so much for that hypothesis.


These laws are probably more about giving a handout to personal injury lawyers than they are about protecting anybody.

Okay the number of accidents rises, but that could be because of increased traffic, road conditions being changed, or individuals indulging in more risky behavior (no cellphones => more time to do possibly more stupid things).

At the very least, one could run a reasonably simple econometric model to sort some of these issues out.

I'm SICK of all these laws in California.

First, they took away my incandescent light bulbs; then they took away my plastic bags; then 52% of the voters--including 92% of African Americans--goose stepped to the polls to take away my right to marry; then they took away my right to text and drive.

In a week, the same majority that goose-stepped to the polls to take away my right to marry will most likely vote to raise my marginal tax rate by 3 points, making my California tax rate 14.3%. And that's when I move to Nevada. We've already moved most of our business.

The cartoon representation of a really bad idea is a guy with a thought cloud over his head and a flourescent light bulb.

Like the toilet, I'm convinced the government went looking for things that weren't broke in order to break them.

The law was enacted at a time when cell phone use and texting was increasing dramatically

I think we should pass a law that bans robots from texting too, unless it is a multi-processor robot.

When we were in Amsterdam recently, we were almost hit by a texting bycyclist who steared with one hand and texted on her phone which was on a mount placed on her bicycle handle.

I mean what are we looking at:

On the one hand, lawmakers responding to whailing family members of the "teenager who killed themselves and all their friends, while texting-see you in 5 mins- to their sibling/friend/parent" with what their paid for, you know, write laws.

On the other hand, statistical anlaysis showing an "odd" counterintuitive response to preventaive harm laws.

Guess who wins. 80% of our laws are passed this way, many of them supporting the biases of MR readers, but without generating shock and horrow in the comments section.

Comments for this post are closed