The Most Dangerous Machine

Steve Levitt is my Freakonomics friend and co-author. He teaches economics at the University of Chicago.

Stephen DUBNER: So Levitt, how good or bad a driver are you, scale of one to ten?

Steven LEVITT: I am one of the worst drivers on the planet. I’m a terrible driver, awful.

DUBNER: Give me a number, one to ten. Two?

LEVITT: One, one.

DUBNER: So what does that say about driving in the United States in the year 2013, however, that you’re one of the worst drivers who ever lived and yet you’re still alive?

LEVITT: Well, we have great technology. So when I bang into things neither I nor the things I hit get hurt, very badly. And we have great developments — like people have very loud horns that they blow at me. And they have really very vivid brake lights and other things that flash in front of me telling me I have to stop. And um they have built technology on the sides of roads so that when I’m veering off my lane towards the grass it goes du-doomp-du-doomp du-doomp-du-doomp-du-doomp. And reminds me to back on. So I think there are a lot of safeguards in there to keep me from killing myself in a car.

DUBNER: Why do you think you are such a bad driver, Levitt? You’re relatively good at some other things.

LEVITT: I’m very distracted as a driver. I don’t pay very much attention. 

DUBNER: What distracts you? Is it stuff in your head? Are you reaching for the hamburger or the music? What is it? 

LEVITT: Partly it’s in my head. And I just am thinking about other things. But I am an incessant channel changer on the radio. I am always in search of a better song than the one that’s currently playing.

DUBNER: Is there any song you’ll stay ‘till the end? Like a Kelly Clarkson? I know you like the girl pop.

LEVITT: Yeah, lately, we’ve been staying to the end of um, when I am with the kids, to the end of “Wrecking Ball,” with Miley Cyrus. They are a big fan of that … And I talk on the phone and I text. I do all the things you’re not supposed to do. I’m pretty much a walking time bomb when it comes to driving.

*      *      *

Hello there. What are you doing while you listen to the podcast? Maybe you’re on the computer, pretending to work. Maybe you’re taking a walk in the park, or you’re out for a run. Or maybe you’re in your car. With your sound system and your GPS and your cup holder – all the comforts of home, at 60 miles an hour. For more than a century, the car has been an extraordinarily convenient way to get around. It changed the way we think about distance. For many of us, the car is an irreplaceable utility – and a vanity. We love our cars. We might even imagine that they love us back. But here’s one more thing about the car: it’s a deathtrap.

Ian SAVAGE: Automobiles in general are probably the most dangerous machine we interact with.

Ian Savage is an economist at Northwestern who studies transportation safety.

SAVAGE: If you look at the only other one which would come close would be firearms, where if you include together homicides, accidents, and suicides, you’re probably getting numbers which are up in the same sort of 30,000 range.

That’s 30,000 as in 30,000 deaths a year, in the U.S. alone, from motor-vehicle crashes. Last year, the exact number was 33,561. Worldwide, there are about 1.2 million deaths. A year. Now, what is it that causes all those crashes? That’s what we’ll be talking about in today’s program. One thing I’ll tell you right off the bat is that more than 90 percent of those deaths are the result of driver error. It doesn’t help that people like Steve Levitt are out there sending texts and looking for Miley Cyrus on the radio. Recent studies show that texting while driving roughly doubles your chance of crashing. Distracted driving would seem to be an obviously bad idea. But before we go forward, let’s go back to when the car was new, to the late 19th century. The cars looked more like big bicycles than the cars we drive today. And the roads were essentially horse paths – they were rutted, muddy, and clogged with manure. And back then, it was the cars that were the distraction. Lawmakers in the U.S. and the U.K. tried to ensure that the car didn’t cause too much trouble:

SAVAGE: Back in the early days of automobiles, when people thought this was sort of going to scare the cows and uh the local populace they for a while actually required people to walk in front of this newfangled machine carrying a red flag so the vehicle could proceed at sort of snail’s pace and you’re being warned that something new and unusual is coming.

Carole LAMBERT: There were people who said that if God wanted people to get around in cars he wouldn’t have made horses.

That’s Carole Lambert. Her great-grandfather, John Lambert, was not one of those people. He is known as the inventor of the first gas-powered car in America. And, not too coincidentally, he is also known as having caused the first automobile crash.

LAMBERT: You know, there’s a marker in Ohio City marking uh where that first accident took place.

DUBNER: What does that marker say?

LAMBERT: It says, “site of America’s first accident.”

Here’s what happened:

LAMBERT: So, John was sitting high on the seat of this car, and he had a passenger with him. And he knew darn well that the streets were rutted and also there were tree stumps that dotted the roads, and roots were protruding. I’m kind of thinking he was talking to his buddy there, and the protruding root turned that little front wheel, and the car went peppering into the side of the road, where there was a horse’s hitching post.

DUBNER: Oh, that’s very metaphoric somehow that the car hits the horse’s hitching post. It’s like the future is trying to knock the past further into the past.

LAMBERT: Oh that’s good Stephen, I hadn’t thought of that. That’s good.

John Lambert and his friend survived the crash just fine. But today, more than a century later, we are all accustomed to the huge toll of traffic fatalities – remember, more than 33,000 a year in the U.S. alone. Over ten years, that’s more than 300,000 people killed – nearly 1 percent of the current population. Traffic accidents are one of the leading causes of death, and easily the top cause of accidental death. While the Great Recession curtailed driving a bit, and there’s talk that we may have reached “peak motorization,” there are still more of us driving more miles than ever before – with more and more potential distractions too: our mobile devices, our lattes, our noisy kids, our podcasts. But what if I told you that we’ve actually been getting safer? True story. Remember how Steve Levitt was telling us what a terribly distracted driver he is:

LEVITT: I am an incessant channel changer on the radio. And I talk on the phone, and I text. I’m pretty much a walking time bomb when it comes to driving.

DUBNER: What you’re describing are all these distractions. Many of which are modern ones. And then there are all these other technological and engineering and design improvements which are also modern. So does this mean that the improvements are winning out over distractions? Do you think that all the driving distractions that people worry about so much now are being kind of washed away by the fact that we’re getting better at keeping people from killing themselves in cars?

LEVITT: Well, the data certainly support that view. If you look at the path of fatalities on the roads per mile driven, it’s an amazing success story.

And here again is Ian Savage, the Northwestern economist:

SAVAGE: The rate of fatal crashes per mile driven, automobile mile driven has declined by two-thirds since 1975.

DUBNER: Wow. And just so people understand there’s a big difference obviously between rate and amount. So there are more people in the country and we drive more than we did in ’75, so if you look at just the raw numbers, the number won’t have fallen as much but you’re saying that the rate has fallen by about two-thirds in the last 40 years, yes?

SAVAGE: That’s correct. In terms of a total count, we were up at just over 50,000  about 1980 and now we’re heading in the low 30,000s.

DUBNER: Wow, so that is amazingly good news, yes? I mean, we can’t deny that.

SAVAGE: Oh, no it’s excellent news, yeah.

So how did we get here? How did we get to the point where the rate of traffic deaths has fallen by two-thirds in the last few decades? Well, clever people have been thinking about how to make auto travel safer pretty much since the beginning of auto travel. Now, let’s be clear: there was a lot of trial and error along the way. First, let’s consider a few of the errors. Around the turn of the 20th century, there was a “cow-catcher” on cars, a fender meant to scoop animals off the road. Didn’t work so well. Later there was a tire covered with suction cups, meant to improve grip – which was a failure. And the auto-entrepreneur Preston Tucker, who had many great car ideas, also had this one, as described by automotive writer Brett Berk:

Brett BERK: He designed what they called a crash compartment. Which was a padded area under the dash on the passenger side, where riders would crawl when a collision was imminent. Sort of an automotive bomb shelter. The idea being I guess you’d have the wherewithal to jump out of your seat and jump into this padded crash compartment in the event of an accident.

Okay, but there were a lot of good ideas too. In 1917, for instance, there was a doctor in Indio, California, named June McCarroll who got sick of treating car crash victims. Here’s Alison Federik, a museum curator in Indio:

Alison FEDERIK: She ended up being a bit of a celebrity because when they first started getting cars in that area – which, the roads were predominantly dirt roads at that time – there were no lines on the road to separate it. There were so many car crashes, and people flying out of their cars — it was sort of a pandemic problem. So she decided to paint a line down the areas of the dirt roads that were around Indio and nearest to her.

Having a centerline on a road will cut crash frequency by at least 20 percent. So, thanks, Dr. McCarroll. Even better than a centerline is a centerline with a rumble strip – which is one of the Federal Highway Administration’s nine proven safety measures, along with roundabouts and safety edges. But wait, there’s more:

SAVAGE: A surprisingly large proportion of highway fatalities occur when there isn’t even a collision with another vehicle, so it’s when a vehicle leaves the highway without prior collisions, over half of the fatalities.

That’s Ian Savage.

SAVAGE: And obviously what you’re doing there is you’re either rolling off down the embankment or even something worse, you’re colliding with a fixed object, usually a bridge abutment. So I think there’s been a lot of developments in crash barriers…where you are improving the highway but it isn’t exactly the highway surface itself.

So all of these road improvements, any one of which is relatively minor, have combined to save a lot of lives. Then there’s the seat belt, one of the most important auto-safety inventions yet. The seat belt came along in the 1950s but it took a long time for them to catch on. Carmakers didn’t like to promote the seat belt because they worried that a seat belt implied that cars were dangerous. Which they were! Especially back then, when car interiors were made of much harder materials than they are today. Until the 1980s, seat belt use was only 10 or 15 percent; today, we’re up to about 86 percent. Which is great news, because a seat belt reduces the risk of death by as much as 70 percent – and, at about 25 bucks a pop, they’re one of the most cost-effective lifesaving devices ever invented. There’s one life saved in the U.S. for every $30,000 worth of seat belts installed in cars – and compare that to about $1.8 million per life for an airbag. Ian Savage has some other ideas about why the auto death rate has fallen so much, especially since the 1970s:

For starters, Savage says, there’s a lot less drunk driving, thanks to tougher laws and social pressure. Over the last ten years, alcohol-related traffic fatalities have fallen by 28 percent.  Also, driver demographics have changed. Younger drivers tend to be more dangerous, and there are relatively fewer of them:

SAVAGE: I looked at the 1980 census and saw that 18 to 29-year-olds were 30 percent of the population. By 2000, that number was down to 22 percent, which is what it is now. So in other words, the baby boom has passed its way through, shall we say the risky driver stage, so I think this is a really important aspect.

Another factor, says Ian Savage, is increasing urbanization. Driving in a city might seem dangerous because there are a lot of other cars and people and buildings and whatnot. But rural areas – those wide-open stretches where you can go a lot faster and hit a tree or another car – are in fact three times more dangerous. So, with more people living in cities – fewer traffic deaths. And then there are cellphones. That’s right: cellphones. Ian Savage tells us that cellphones, as distracting as they would seem to be while you’re driving, may also be responsible for saving lots of lives:

SAVAGE: You always hear the medical professionals talking about this whole golden hour after the trauma occurs of getting high-quality medical care. So it’s easy probably for you and me to remember not that long ago if you had a crash in a rural area, boy how would you go and phone to tell someone this has happened. You’d have to get passing motorists who would have to drive to the nearest town or something like that. So clearly there’s that aspect of things. So I mean I think that my observation of the cell phones and safety literature is that it’s controversial. There are findings both ways in terms of the overall effects.

So cellphones may drive auto fatalities down, in that they let us call for help after a crash. But as he acknowledges, that benefit may be outweighed by the cost of distraction:

SAVAGE: I do see that things have kind of morphed in the last 10 years. Ten years ago people were sort of worried about drowsy driving, which I think we still are concerned about, right. But I think it’s morphed into this whole kind of distraction.

So when we come back, you’ll hear about driver distraction in many forms:

Greg FITCH: You know, we have evidence of people talking on two phones at once, while smoking a cigarette while driving through a work zone and running a yellow light. We’ve seen videos of drivers relieving themselves behind the wheel.

And we ask what happens to driving once our cars learn to drive themselves:

STRAUS: It will no longer be about the utility of getting from point A to B with automobiles. There will be places like this around the country kind of like horse ranches where you can still enjoy the sport but that’s what it’s about, is the experience, not about getting somewhere.

*      *      *

We are talking today about “the most dangerous machine” – the automobile – and what’s made it less dangerous over time. But also the factors that do make it dangerous. Distractions, for instance. All the things we do while driving that we probably shouldn’t be doing. But here’s a question for you. When researchers are trying to figure out just how dangerous a given distraction may be, how do they know? They’re obviously not sitting there in the passenger seat with distracted drivers. Or are they?

FITCH: My name is Greg Fitch. I’m a senior research associate at the Virginia Tech Transportation Institute.

Greg Fitch’s outfit is a leader in what’s known as naturalistic driving studies. So, no, they don’t ride along with drivers. They put a video camera in their cars. Now, I know what you’re thinking – what about the observer effect! Won’t a driver who’s being taped always be on his best behavior. Greg Fitch says no.

FITCH: People soon forget that they’re actually being recorded. And the way we’re able to determine this is we see them do all kinds of ridiculous things that no one, no one in their right mind would do if they’re being videotaped unless they’re on a reality TV show. You know, we have evidence of people talking on two phones at once, while smoking a cigarette, while driving through a work zone, and running a yellow light. We’ve seen videos of drivers relieving themselves behind the wheel. And for the over-the-road drivers, we’ve seen examples of them actually meeting with a prostitute behind the wheel, you know, so these are things that you don’t actually see in an empirical laboratory study. This is true driver behavior that we’re capturing.

Okay, so when you can watch drivers driving, and try to measure the effect of various distractions, what do you find? One conclusion: talking on the phone doesn’t seem to be a very big deal. Fitch says that people talk on a cell phone about 10 percent of the time they’re driving. And he and his team found that, whether a driver is holding the phone or not, that talking on the phone – while not recommended – does not increase the risk of a crash. In fact, it can even be associated with a positive outcome:

FITCH: It’s been proposed that talking on the phone reduces boredom. There’s been research that shows that people are more awake, more alert up to 20 minutes after having a cell phone conversation. Kind of like, you know, drinking a cup of coffee or taking a caffeine pill.

And if that’s not surprising enough, how about this:

FITCH: A big theory that was proposed over the last two years is that people may actually be adapting or changing the way they drive while talking on the phone to kind of mitigate the risk. 

In other words: we seem to be getting better at multitasking. At least when it comes to driving and talking. But what if you stop talking on your phone and start using it for something else – like texting, or checking your fantasy football lineup? That is bad news.

FITCH: Anything that’s dangerous behind the wheel all have a commonality, and that is when you take your eyes off the road you’re at an elevated risk. It’s been shown time and time again. Reaching for the phone, dialing, texting, even putting the phone away are associated with an increased risk. Whether eating a burger, whether you’re putting on makeup, you know, those all have commonalities and that is you should not take your eyes off the road when driving.

So there’s the dilemma. We are so dependent on our cars, at least for the foreseeable future – but, when driving them, we are our own worst enemy. It’s nice to know that driving has gotten relatively safer – but remember, there are still 1.2 million traffic deaths a year worldwide! Not to mention 20 to 50 million injuries and a financial cost of roughly half a trillion dollars a year. So what if we look at the slightly less foreseeable future. What if we look at the driverless car? No one can say exactly when or how it will happen but it could well be that seven or ten or 15 years from now, most cars have the ability to drive us, rather than vice versa. A lot of the necessary technology is already in many cars – like GPS and onboard computers and sensors and cameras. And just about every major carmaker, along with Google, is testing driverless cars, with much success. There will be a lot of wrinkles to work out, of course. But in theory, a computer-operated car will be much safer than a human-operated car. It won’t drive drunk, or while putting on mascara, while trying to open the packet of ketchup. There will of course still be accidents. Fred Cripe worked for Allstate Insurance for 31 years. He’s now a consultant:

Fred CRIPE: I think in the future, what you’ll find is when vehicles are involved in accidents it will simply be an ‘OK, the guidance system didn’t work in this case.’ And whoever’s guidance system is in the vehicle pays for the loss.

I asked Ian Savage, the transportation economist, to try to predict some other advantages of the driverless car:

SAVAGE: When you talk about older drivers and diminished driving skills amongst older drivers, the autonomous vehicle could change their lives tremendously and also improve the safety of the pedestrians who have to walk around these elderly drivers driving on the streets. I think the biggest gain is not going to be safety-related, it’s going to be in road space and the issue of congestion, because if we can get by the issue of the fact that on most highways there are large gaps between automobiles, hence a lot of wasted space on the highway, if we can now have vehicles moving faster and closer together we can make a big dent in urban highway congestion.

Not long ago, I had a chance to ride in a driverless car – it was a Cadillac SRX4 that engineers at Carnegie Mellon University in Pittsburgh have been developing with General Motors. One of the engineers sat in the driver’s seat but he didn’t drive. He just put the car in autonomous mode and then we rode along. 

We weren’t out on the streets with live traffic – although Carnegie Mellon engineers have done that. We rode around a big street grid that the university built on the site of an old steel mill. They set up impediments – bicyclists, pedestrians, other cars, construction barriers – and in each case the Cadillac recognized and dealt with each one safely: steer, brake hit the turn signal, everything a human driver does except check for e-mails or look for Miley Cyrus on the radio dial. It was an extraordinary experience – but mostly because it felt so ordinary: there was nothing “futuristic” feeling about riding in a driverless car. In fact, it was pretty boring after the first ten or 15 minutes. Especially if you’ve ever ridden in, say, an airport shuttle train that’s already fully autonomous. Or an elevator. Technological marvels come along; we scoff at them, say they can’t possibly work, and then when they do, we immediately forget our doubts and start looking for the next marvel. And what if the driverless car then really does take over? Will it be much, much, much safer than how we travel today? Probably. But for some people – especially people who love to drive – that may not sound like a great trade-off. Safe is good – but boring is bad. How will future drivers get to experience the thrill of driving in a driverless world? Well, to get a taste of that, we visited a place that already exists. We leave you with our visit to the Monticello Motor Club in upstate New York where driving isn’t just something you do to get from here to there. 

Ari STRAUS: I’m gonna steal your car for a minute! Why don’t we hop in the car and drive around a little bit? So we’re going to go ahead and start up the 2013 Jaguar F Type convertible.

STRAUS: My name is Ari Straus. I’m one of the partners and the managing member at Monticello Motor Club which is an awesome private country club racetrack about 90 minutes northwest of New York City. I’m really an IT geek, my background is mostly startups in the healthcare and software field. But I’m also a huge enthusiast of cars, I’ve been racing for several years.

STRAUS: Now let’s head up to the north course and see what it’s like to take a lap in this.

STRAUS: What I love about driving on a racetrack is that it’s a sport that requires 100 percent concentration. It’s adrenaline combined with the best of meditation. You can’t think about anything else when you’re on the racetrack.

STRAUS: We’re driving up now through the north pits. We can also run the north configuration. It’s a 1.9-mile twisty course. Simultaneously with our south course which is 1.6 miles.

STRAUS: We’re entering the north straight here. Let’s give it a little more acceleration. 

STRAUS: Ok, what you hear is 495 horsepower v8 that’s uh pushing us up to 140 miles an hour on that straight. As we’re going around the racetrack the first thing I feel is being pushed into the back of my seat every time I push the accelerator. Then when we come around the turns, I’m being thrown to the side. Now, the seat is incredible so it’s holding me in place.

STRAUS: That’s the engine just, uh, telling us it’s alive and well.

STRAUS: The safety of modern cars and modern racecars is a completely different world than even ten years ago. You’re surrounded by airbags, you’re surrounded by pillars that are gonna protect us whatever happens to the car. A roll hoop behind you, an aluminum hoop that’s meant to hold all the weight of this car if it were the shiny side up. Belts, restraint systems, even the contours of the seat, how we’re held in here. And if there’s an impact well, usually what’s hurt is your ego, but not your body.

STRAUS: In 20 years, this will be one of the beacons of experience for combustible cars. For people who drive for the thrill of driving, but it will no longer be about the utility of getting from point A to B with automobiles. There will be places like this around the country, kind of like horse ranches where you can still enjoy the sport but that’s what it’s about, is the experience, not about getting somewhere.

STRAUS: Did I steal your car? 

STRAUS: O.K. Not much faster. A little bit faster. We can grab some lunch inside.