Freakonomics Radio - 548. Why Is the U.S. So Good at Killing Pedestrians?

Episode Date: July 6, 2023

Actually, the reasons are pretty clear. The harder question is: Will we ever care enough to stop?

We made an episode almost a decade ago called The Perfect Crime. The idea was that if you wanted to kill someone and not go to prison, the best way would be to simply run them over with your car. That's the way it works, especially in America. If you're driving a car and you kill a pedestrian, even if it's entirely your fault, most likely nothing very bad will happen to you. In our legal and transportation systems, the car is supreme and pedestrians don't have much protection. Back when we made that episode, I guess we had a slight hope that presenting the data and discussing the problem might lead to some progress.

It did not. In 2014, the year we published The Perfect Crime, there were just under 5,000 pedestrian deaths in the U.S. Last year, around 7,500 deaths. And it's not just the raw numbers increasing. The rate of pedestrian deaths per mile driven is the highest in more than 40 years. Even during the pandemic, when driving was way down, pedestrian deaths continued to rise. Among the world's high-income countries, the U.S. is particularly good at killing pedestrians. The death rate here is much higher than in places

like northern and western Europe, Canada, and Japan. So today on Freakonomics Radio, we're back at it with one simple question. Why? Why are we a world leader in this terrible statistic? The cars we're driving are bigger, fatter, faster. The problem of distraction has gotten much worse. In the United States, we've decided that car movement is really the supreme consideration when it comes to designing our streets. Okay, I lied. It isn't just the one question we want to answer. We also want to know, what are the best ways to stop the killing. This is Freakonomics Radio, the podcast that explores the hidden side of everything with your host,

Stephen Dubner. As you know, or at least I hope you know, I try to be an honest broker on this show. I'm not trying to be a pundit or an activist or an opinionista. What I try to do is talk to well-informed people and find data that sheds light on issues that often lie in the shadows. But in this case, let me just say, for the record, I am 100% against the killing of people by cars. Is it true that pedestrians are sometimes at fault? Of course, but mostly not. And even if so, you could argue that the penalty for stupidity should not be death.

If it were, we'd probably have a much smaller population today than we do. I am especially against the death of pedestrians because it is so plainly not a fair fight. I live in Manhattan, and when my kids were little learning to get around the streets, here's what I used to tell them. When you are a pedestrian, you can do absolutely everything right and still end up dead. Here's an interesting statistic to consider from NHTSA, the National Highway Traffic Safety Administration. Only 2% of all people injured in traffic crashes are pedestrians, but pedestrians account for 17% of all traffic deaths. That's a pretty good sign that when pedestrians compete against cars, they are bound to lose. Some of the fault, as we will hear today, lies with drivers themselves.

But let's start with a wider view. I want to be clear. Drivers are operating in a system that is designed to create crashes. That is Kelsey Ralph. She's a transportation scholar at Rutgers University. We demand high speeds. We want to get to the places we want to go quickly, in part because Americans drive so many miles a day. The average American spends 55 minutes a day in their car. And we want to get to wherever we're going quickly. So we have high speeds. We

have a lot of cars on the same roadway. And then we have what we call arterial streets. So imagine two or three lanes in each direction, a McDonald's, a Home Depot, lots of driveways. Maybe there's sidewalks, maybe there's not. People are pulling in, people are pulling out. You're going to have crashes there. And in fact, when we do crash hotspot analysis, these are by far the most dangerous streets in the nation. In most cities, about 80% of the crashes occur on about 15 to 20% of the roadways. In New York City, where 119 pedestrians were killed last year,

one analysis shows that just under 7% of the streets account for more than half of all pedestrian deaths. That analysis comes from this man. Philip Miatkowski, and I'm the Senior Director of Research and Policy at Transportation Alternatives. And Transportation Alternatives is what? Transportation Alternatives is a nonprofit advocacy group here in New York City for 50 years now, whose mission is to reclaim space from cars for bus riders, cyclists, and pedestrians. So Transportation Alternatives is an advocacy group, plain and simple, and some people view them as devoutly anti-car.

But let's hear some of the data. Using New York City's open data, we calculated that 76% of New York City's street space is dedicated for either moving or storing vehicles. 23% is dedicated for sidewalks, and less than 1% is dedicated for bus lanes and bike lanes combined. So the goal here is just to rebalance the streets to better reflect how New Yorkers already move around. So we happen to be sitting in a recording studio in midtown Manhattan. Are any of these dangerous intersections pretty close by? Can we go take a look and try to avoid getting killed?

Yes, absolutely. Nearly every avenue near the studio here is one of those 6.98% of New York City streets. So you're saying the minute we walk out of the studio, if we go left or right, you're saying in either case, we're coming into a kind of pedestrian danger zone? Yes. Let's go. So we're on West 45th Street, approaching 10th Avenue, one of the more dangerous streets in New York City.

You take me to all the nicest places. So the street here is really wide, I would say similar to a highway width. Okay. And describe this and how it's, I mean, it's pretty obvious even to me, it looks a little bit like the Indy motor speedway, but just the straightaway part. This is a street that's mostly dedicated to funnel as many cars as possible at the most comfortable and fastest rate for them. So most of these cars are at or above the speed limit. They have four to five travel lanes. There's no space given to cyclists, no space given to bus riders,

even though there are a lot of buses on this route. So when you combine all of those factors, it creates this dangerous environment here. There's probably, just eyeballing it, 1 30th or 1 50th, the amount of pedestrians here than there were just a couple blocks east, which is Times Square and bordering Times Square, you think? Correct. So what if I argue, you know, what's wrong with taking an area with low pedestrian density and making that wide open for cars and in the areas that are heavily populated by pedestrians,

treating it totally differently. Yeah, well, the street that you design for, you're going to get the uses that the design is inviting. So there are fewer pedestrians here, maybe because they do feel less space. The sidewalk is narrower. And also, you know, who are we building our streets for? Is it for the commuters who live in a different state, in New Jersey or in Westchester County? Or is it for the residents of New York City? I do recognize that we're always going to need cars in society,

whether you're disabled or whether you live far out from the city, or just because you want to. But right now, there's such a huge imbalance. 76% of our space is dedicated for cars when the majority of New Yorkers don't own a car. Do you want to cross some streets? Sure. Okay, so we have the light here. We're crossing north to south on 45th. Everybody's behaving here.

Nobody's running a light. So far, so good. We made it across 10th Avenue, no problem. But the data back up Mietkowski's claims about the danger here. Over a recent two-year period, one stretch of 10th Avenue, running from 14th Street up to 59th Street, that's just over two miles, saw three pedestrian deaths and more than 150 injuries, as well as one cyclist death, nearly 100 cyclist injuries, and more than 200 motorist injuries. So before we continue our adventure with Mietkowski, it's worth taking a step back and asking,

how did our streets get so dangerous in the first place? And why, over the past 20 years, have pedestrian deaths risen much more than other traffic deaths? So many reasons. That, again, is the transportation scholar Kelsey Ralph. The cars we're driving are bigger, fatter, faster. If you get hit by a sedan, you break your legs and you get thrown over the windshield. If you get hit by an SUV, you get hit in the chest and then you get pulled under the vehicle. So much more likely to die. And we're selling way more SUVs than we used to.

We're driving more. We're also walking more. But pedestrian deaths per million miles driven increased 30% during that time. So it's not just more driving and more walking. It's something different. I think distracted driving remains really widespread. And interestingly, when we looked at police enforcement of distraction,

every agency we were talking to said, oh yeah, we do a lot of distracted walking enforcement. We do a lot of ticketing of people walking distracted. Not so much of people driving distracted because it's really hard to see them. Also, let's talk about regional location in the United States. Americans, they're moving to less safe regions. People are emptying out of the Northeast, which tends to be fairly safe for people walking and biking, and moving to places like Florida and others that are really

high on the pedestrian danger index. Let's review the four reasons Kelsey Ralph just gave for the rise in pedestrian deaths in the U.S. Number one, vehicles are bigger and more dangerous. Number two, there's more driving and more walking. Number three, drivers have more distractions. We'll get into those details later. And number four, people are moving away from older cities where narrower streets decrease car speeds and make it safer to walk to places with wider, faster streets. There is at least one more reason why so many pedestrians keep getting hit and killed in the U.S. So most states have legislation that requires engineers to use what's called the 85th percentile rule when setting their speed limit.

OK, and what is the 85th percentile rule when setting their speed limit. Okay. And what is the 85th percentile rule? For that, let's hear from Sarah Bronin. So the 85th percentile rule is a rule in the Manual on Uniform Traffic Control Devices, which is that federal publication that is published by the Federal Highway Administration. Bronin is a professor of law and urban planning at Cornell. In 2021, she and Gregory Schill published an article in the Harvard Law Review Forum called Rewriting Our Nation's Deadly Traffic Manual. That's the 818-page manual on uniform traffic control devices for streets and highways, also known as MUTCD. And it says that for any given roadway, the speed limit can be changed if 85% or the 85th

percentile of drivers are driving a certain speed, even if that speed is higher than the suggested speed limit. So what it does, unusually among laws generally, is it actually allows a law, the speed limit for a particular road, to be rewritten by the lawbreakers. The idea is that the 85th percentile rule recognizes what everybody would view as the commonly acceptable speed of traffic. But what it really does is it gives the ability to make a road faster, and it hands that ability over to people who were recklessly driving in the first place. Okay, so I don't mean to be naive about the ways of rulemaking and lawmaking, but why on earth is that still on the books? Well, in a recent opening of the MUTCD for its 11th iteration.

Opening meaning an opportunity to revise, yes? Exactly. So the Federal Highway Administration put out a draft of the MUTCD and invited public comment. And there were thousands of comments that encouraged the Highway Administration to end the 85th percentile rule, primarily to make roads safer. The issue is that drivers drive the speed that the roads tell them to drive. So the 85th percentile rule is a problem, don't get me wrong. But it also illustrates the fundamental issue, which is that we are designing our roads to enable cars to drive faster in a way that disrespects the other users of a roadway. Bronin says there's another problem created by these federal traffic guidelines, a problem

that may help explain why drivers who kill pedestrians are rarely punished. If you design streets to those standards, as a municipality in particular, you don't have legal liability because you've met this national standard. By adopting these standards, what happens is you preclude, especially municipalities, from the kind of experimentation that can improve streets and that can tell us which streets work best. And yet you see local governments like New York City and others aspiring to make their streets safer, even in the face of these external constraints. New York City has adopted an approach called Vision Zero, which was pioneered in Sweden. It is intended to entirely eliminate traffic deaths and serious injuries. Here again is Kelsey Ralph.

I have data from Edmonton, Canada about how many people get hit by drivers. Drivers hit 284 people there in 2017. And so we could really zoom in and look at the actions of the people who were hit. What were they wearing? When were they crossing? Where were they crossing? But if we zoom out, we see that drivers in Edmonton hit 49,000 other objects that year, other than pedestrians. And 5,500 of them were stationary. We're talking about parked cars and signs. These aren't things darting out. It's tempting to focus on the pedestrians because that's where the action is in terms of deaths. But I think we need to focus on the drivers because that's where the action is in terms of risk. Okay, so let's go where the action is. Coming up after the break.

The gorilla beats his chest and walks out the screen. About half of the people don't see it. That's exactly the kind of impairment that cell phones produce while driving. I'm Stephen Dubner. This is Freakonomics Radio. We'll be right back. One of the trickiest things about understanding pedestrian deaths is the lack of data on what happens right before a crash. As you can imagine, the driver who hits a pedestrian may tell a story that takes the blame off themselves, and the dead pedestrian isn't around to tell their story. So yes, tricky data to get, but also important data to get. Which brings us to this man.

My name is David Strayer, and I'm a professor in the psychology department at the University of Utah at Directly Applied Cognition Lab, and we study driving and driving behavior. Strayer has come to know a lot about driving behavior, especially bad behavior. When we look at fatalities on the roadway, the four things that are killing people are speeding, alcohol and intoxication, fatigue, and distraction. Distraction has been prevalent for a long time, but increasing quite rapidly, and we think that that's one of the major reasons for the increase in the number of roadway fatalities. Much of what Strayer knows comes from the work he does in his lab.

We have two advanced driving simulators in my lab. I have five on-road vehicles, and we test people between the ages of 16 and 92. When it's safe to do so, we'll test them on the road. When we're engaging in some kind of activity that we think is too demanding, we would do it in the driving simulator. And just list for me some of the distractions that you're introducing, whether it's in the simulator or in the vehicles on their own. Listening to a radio, which on our one to five scale is like a 1.2.

You can measure it, but it's not particularly distracting. Listening to a book on tape has a score of about 1.7, so more because you have to pay more attention to the audio content. And by the way, five on that scale was the most mentally challenging task that we could think you could do, kind of like balancing your checkbook while driving down the road. And so that five is a really high watermark. And you should really think carefully anytime a task gets anywhere near that. Talking to a passenger or someone on a cell phone has about a 2.2, 2.3. And interacting with these voice assistants that come along with your phone, that's oftentimes

four or higher. It can get a little bit better with perfect systems. We've been able to develop some in the lab that are perfect and don't make errors. But the real world systems that we all really use on a day-to-day basis, it's probably a four or higher on that five-point scale. I would think that using voice commands for a phone or for the car's dashboard would be less distracting for a driver than using their hands for the same tasks. Is that the case? It turns out that using your voice is not, there's no free lunch there. It takes brain power to think about what you're going to say,

how you're going to say it to that voice assistant. Oftentimes that voice assistant gets it wrong and does the wrong thing, and it takes longer. Some of these activities take 20, 30, sometimes 40 seconds to fully complete. If you're trying to, say, adjust the temperature in your car, you can just reach over and push one button, and that's a much shorter interaction. So, time on task is really important, and the voice assistants tend to go slow and are error-prone. This distraction represents the cost of what the economist Daron Asimoglu calls so-so technologies. One of his favorite examples is the self-checkout counter

in grocery stores. Yes, they work, sort of, sometimes, and eventually they may work reliably. That's the path that successful technologies tend to take. Voice assistants will hopefully take this path too. Although, if we take a step back, the bigger victory here would be a different and much more complicated technology, autonomous travel. Humans are good at many things, but when it comes to driving, we are extremely fallible. The problem is autonomous travel isn't here yet, which means we are stuck with humans driving for now. And one thing that makes humans so fallible, according to David Strayer's research, is

just how easily we get distracted. One of the things that happens is something called inattentional blindness, where something happens right in front of you and you just don't see it. It's a look-but-don't-see type of hazard. When you're using a cell phone, it doubles the rate of that look-but-don't-see phenomena. What does that mean, inattentional blindness? Can you draw a picture of that for me? Sure. Normally, you could think about if I'm looking at something, I could see it. It would register right away. But it turns out that you require attention to be able to process that visual image. When you start to multitask, you take attention away from visual processing of the driving scene. And so what happens is that you actually don't see things

that are right in front of you. Your eyes are on it. Your eyes might be looking right at a pedestrian, right at a car, right at a traffic light, and you zone out and don't see it. We've done brain imaging studies where we can actually show that the brain just doesn't register that object that's right in front of you. Is the famous invisible gorilla experiment an example of this? It is. The work that Dan Simons did with the invisible gorilla is a good example of that inattentional blindness. In that example, a gorilla walks onto the scene right between people passing a basketball back and forth. The gorilla beats his chest and walks off the screen. About half of the people don't see it. That's exactly the kind of impairment that cell phones produce when a driver is using that cell phone while driving. In terms of Psych 101,

the reason that multitasking is compromising driving is that you're taking brainpower away from the driving task. Let's say the average person says that they are eight on a scale of 10 successful at multitasking. How successful is that person actually at multitasking? So first, I'd say that they probably don't have a good assessment of their actual ability. We've tested tens of thousands of people across the world in some tasks that look for super ability. We call those folks super taskers. The statistics say about two and a half percent of the population are in that extremely proficient multitasking category, but 97 and a half percent of us are not. And unfortunately, the people who think they're most likely to be excellent at multitasking

are the least likely to actually be good at it. So do you have any sense of how common texting while driving is? We think that just in observational studies, about 1 in 10 drivers are obviously doing something with their phone while they're driving, holding the phone, texting or something. So it's at least 1 in 10 drivers and probably substantially more. And that's 1 in 10 at the moment of observation, not during the course of an entire trip? Yes. At any point in time, as we're speaking right now, 10% of the drivers are engaging in some kind of cell phone-related activity that's potentially distracting. As distracting as a phone may be, there's another distraction that's built right into

our cars, the dashboard display. There's not a lot of regulation that really focuses on what's in the vehicle. NHTSA has released guidelines for dashboard information systems, but they're voluntary guidelines and automakers rarely follow them. Some of the cars we've tested have 200 multifunction buttons. Some have rotary dials and gesture controls. So there isn't a lot of oversight in terms of what is going on in the vehicle. And for the most part, the consumer, when they buy a

car, has no idea if it's an easy or difficult system to use. Which automakers have the best and worst information systems? Yeah, I'm kind of reluctant to deliver that right now. You can look at some of the papers that we've written, but the point really is not necessarily to call anyone out, but to hope that they all improve because every one of them has room for improvement. Yeah, I do see. I'm cheating. I'm looking at your paper. So the Nissan Rogue has the most demanding, what does IVIS stand for? That's a jargony term for the technology in the car. It's an in-vehicle information system. So that's screens, heads-up displays, the things that support calling, texting, destination entry, navigation, and so forth.

The Kia Optima had the least demanding, and the Kia is not an expensive car. It did make me wonder, is it a lot more expensive to design a better information system? No. I mean, unfortunately, what you see is a lot more expensive to design a better information system? No. I mean, unfortunately, what you see is a lot of times the auto manufacturers will put really sophisticated types of ways of interacting with the vehicle, and they're not fully vetted to make sure that that's actually a good way of doing it. So sometimes simple is better.

Sometimes that simple button press is easier than a multifunction display. Even though it may not look super elegant, it turns out to be engineered well for how we interact with technology. How do you think about distraction as a cause in, let's say, a pedestrian death? And let's start with distraction from the driver. How significant a factor is that? Most of the fatalities are at intersections or places where you have other vehicles that are crossing with other vehicles or other vehicles that are interacting with pedestrians. You see that drivers are much more likely to be distracted at intersections. Maybe

they're trying to send a quick text or talk. We know that talking on the phone has about a fourfold increase in crash risk and texting is much worse. So a driver, when they're multitasking, is putting the public at large at risk. They're putting themselves at risk too, but the greater harm in the last 10 years or so has been pedestrians, bicycles and motorcycles. But drivers in every country have phones. So why would pedestrian deaths be so much worse here? Road design is one big difference. As we've already heard, American roads are engineered for efficiency and speed.

But David Strayer has another thought. It may well be associated with things like the prevalence of manual transmissions in Europe, where they have to be engaged with the driving task. It's hard to text and shift versus in the U.S. where you have an automatic transmission and that supports other kinds of activities. It's a complex landscape, but the risk to pedestrians and other vulnerable road users is actually growing at an alarming rate. And Strayer says it isn't just the drivers

who are more distracted. Every driver has a phone. Almost every pedestrian has a phone. So you have situations where the two parties are both multitasking and consequently neither of them are paying full attention to what their task should be, which is to walk safely across the intersection or drive safely. I live in New York City where pedestrian deaths are shockingly common. I walk a lot and I also love my noise-canceling headphones. So am I a total idiot for wearing them when I'm walking on the street? If you can't hear the ambient traffic, then you're putting yourself at risk, yeah.

Yeah, but you're not quite willing to call me an idiot, it sounds like. Well, let's put it this way. At the end of the day, do you want to get back to your apartment after you've gone on that walk? If you do something where you wear noise-canceling headphones or something where you can't hear ambient traffic, you're putting yourself at risk. You can't hear horns honk, you can't hear cars approaching, and so you can't be as alert and attentive as you might want to be. So there's good evidence that technology can distract us, whether we're driving or walking. Of course, technology can also increase safety in cars, and it has. Kelsey Ralph from Rutgers has another theory that might explain the rise of pedestrian

deaths, even during the pandemic. This has to do with what psychologists call other-regarding behavior, which is a slightly awkward name for how we act toward one another. It's no surprise that people got more angry and a little bit more impatient and a little bit more hostile to other people during the pandemic. And I think that's affected driving. It's also, of course, true that people were driving less. And so there were fewer cars on the road, which meant people could travel more quickly. That is going to make crashes that do happen worse. So I think there's sort of a one-two punch there. It'd be interesting to see as people return to driving, if this decline in other

regarding behavior persists. I suspect it will. You say you think that this other regarding behavior has contributed to the situation, but is there evidence? Not as far as I can tell, no. It's such a shame, but you can see like the rise in people yelling at airline attendants, right? It's all anecdotal and it's all pointing in the same direction. But this is a point of disagreement between me and some other transportation professionals. Some will say this focus on other regarding behavior is a mistake. It's just the system.

And I'm a systems girl. I really want us to improve our infrastructure. But at the end of the day, we need both. Safe infrastructure is not sufficient if the drivers are road-raged. Coming up after the break, given what we've learned about the causes of pedestrian deaths, what can be done to stop them? I'm Stephen Dubner. This is Freakonomics Radio. We'll be right back. You will remember that earlier, Philip Mietkowski of Transportation Alternatives took us to one of New York City's most dangerous intersections. He also took us somewhere much safer, just a few blocks away.

Yeah, this is actually my favorite intersection in New York City. So we're standing on the corner 39th and Broadway in Manhattan. I am actually standing, I think, in the street, but it doesn't seem to matter because there are no cars allowed here. So if you could just describe this block, what we're looking at and how it's different now than it was a few years ago. So actually cars are allowed on this block, what we're looking at and how it's different now than it was a few years ago. So actually cars are allowed on this block, but you see that it looks very different than the typical New York City block. It's painted differently. It has this grayish color. It has granite blocks that slow vehicles as they're making that left turn.

There's lots of space for greenery. And the street bed where the travel lanes are located is in an irregular shape. When you do see a car turning onto the street, they slow down significantly down to five or 10 miles per hour. So how wide is the street from curb to curb here, roughly? It's going to be 40 feet. 40 feet, which as city streets go, is pretty wide, correct? And then of the 40 feet, how many feet are devoted to car travel here then? Well, actually, that's what makes this street so unique.

It's called a shared street. Everything that's designed in the street is supposed to signify what you were under the impression first, that cars weren't allowed. So it's a shared space both for... Back up. I mean, it would be great for our show if the transportation alternatives guy got hit by a car. But for you, Philip, I don't really want that to happen. Yeah, thank you. I appreciate that. The idea of shared streets is not at all new, and they're still common in some cities that were built before the invention of the automobile. Here again is Sarah Bronin. The safest streets in Europe are streets where pedestrians and cars are entirely intermixed.

In other words, where there's not even really a distinguishable sidewalk. You see those streets in England, you see them in Switzerland, where cars, bicyclists, pedestrians are all navigating, sometimes even without striped lanes, the same roadway. And the reason that that is safer is because the car is lacking clear direction as to where their exact lane is and where the sidewalk is. They have to be more careful, so they slow down. The pedestrians who, in such a situation, have as much of a right to the road as any, they're not relegated to sidewalks, can actually feel like they can walk in between cars.

And again, despite our intuition, our probably American mindset, that that kind of situation might be dangerous, it's actually in many ways safer because it signals to drivers that they're not the only users of the road. You can see why shared streets might work well in some contexts, like the dense urban areas Bronin is talking about, but you can see why in other contexts they wouldn't work at all. In fact, some safety advocates argue for the exact opposite solution, essentially quarantining cars from pedestrians and cyclists wherever possible. I asked Kelsey Ralph what she thought of that idea. that drivers have to travel at high speeds, then yes, we should separate people walking and biking. But in so many parts of our cities, thinking about those arterial roads with, you know, the McDonald's and the Home Depot, there are going to be people walking there. There are going to be people biking there. And to the extent that's true, we have to lower speeds to make it safe

for them to mix. So if I were to ask you a basic question, are drivers too protected or maybe too privileged in the U.S.? What would you say? I would say yes. Their needs are prioritized to a greater degree than they ought to be and to a much greater degree than they are in other countries. And meanwhile, back on the corner of 39th and Broadway with Philip Mietkowski. So everything that's designed to this street is signaling to the driver, hey, this street is different. Slow down. There are people, you know, allowed to cross anywhere within the block.

I'd like you to just stand here and literally describe anything you see moving. Yeah, so I see a FedEx guy pushing a cart full of mail. I'm seeing lots of cyclists using the protected bike lane, acoustic bikes, electric bikes as well. Right in front of me, there's a huge pedestrian plaza that takes up an entire block that is full of people and music. Seeing lots of pedestrians here, significantly more pedestrians than drivers. I see somebody on a cargo bike delivering, looks like, water. A unicycle user. An electric unicycle. Yes, electric, yeah. There's vehicles. I see a NYPD cruiser. I see a car turning here left onto our shared street at a much slower pace than a typical street.

So if all of New York City were like this block, how safe would New York City be? We would be pretty close to zero traffic fatalities. And remember what Kelsey Ralph told us earlier about the concentration of crashes. In most cities, about 80% of the crashes occur on about 15 to 20% of the roadways. And so we can make changes to that number. We don't have to change 100% of our roadways. That would be way too difficult. But we can be really strategic and target these places and immediately see benefits in no longer killing our neighbors. And what are some changes that Ralph would like to see? I love speed cameras. Okay, because why? Speed makes crashes more likely, more dangerous,

and less likely to survive them. The best way to slow people down is to make sure we have rigorous enforcement. We can't have a cop at every corner. We can't afford that. But we can have a camera at every corner. And the beautiful thing about it is that it reduces recidivism. People get a ticket the first time the cameras are up, and then they don't get another ticket. Hand is raised here. How about you? Have you gotten one? Oh, me? No. I'm a very careful driver. Well, I am too, I thought. But yeah, there was that one intersection. I don't know if I was speeding so much as I ran the light. But did you do it again?

No. No. What other measures do you think are really effective besides speed cameras? Narrowing the roadway is super useful. So if you look at a roadway and it feels like you should be able to drive 30, 40 miles an hour, you're going to drive 30, 40 miles an hour. And so what we need to do is make the lanes a little bit narrower, add pedestrian infrastructure in the middle, places for people to cross. These things all send signals to drivers that they need to drive at an appropriate rate.

How about speed limit signs? Do they work? The evidence on this is pretty contentious, but it seems to be that lowering the speed limit five miles an hour leads to about a two to three mile per hour reduction. So not bad. 50% return on a little paint. Exactly. Are there any speed reducing or just pedestrian safety measures that people think are effective that in fact are not? Educational campaigns. Ah, tell me about that. And I've seen no evidence that these things are effective. There's a line item and lots of budgets for them. So it's something we keep doing.

It's also much easier to do, right? We did a successful PSA campaign. Yay, we are solving the problem. The other options are really hard. It's really difficult to reduce speeds. It's really difficult to get pedestrian infrastructure. How well or poorly would you say the media describes transport generally, especially road transport? I think on average, the media tends to defer to whatever that quote-unquote experts are saying. I feel horrible for journalists, right? They're working in newsrooms that have been emptied out.

They're on a really, really short deadline. They're just trying to get the story out as quickly as possible. And so as a result, they tend to follow almost word for word police press releases. And that's sort of a nice first blush. But the problem is police press releases have all kinds of victim blaming within them inadvertently. So some colleagues and I, we went and trained police officers in New Jersey and are hoping to expand this more nationally, where we said, look, you're using language that is victim blaming.

You're using complete lack of agency. A pedestrian was hit. By what? By whom? It's unclear from their analysis. And so we actually gave them a template of, here is sort of a well-written press release, and you can fill in the blanks with your crash's details. Traditionally, the coverage would say something like a pedestrian was hit on First Avenue at 10 p.m. Instead, we'd suggest that a driver hit a pedestrian on First Avenue, which is a two-lane road with a 45 mile an hour speed limit, adding this contextual detail and subtly shifting responsibility to the driver of this vehicle. Now, of course, the caveat is that if they found out that this was a drunken pedestrian

who had ran out into the street, we should be responsive to what's happening. But the default up until then has always been to blame the pedestrian with no information whatsoever. And now we're saying, let's shift the default so that we're blaming the driver. And that might be wrong sometimes, but the current default is even more wrong. So given our history, this podcast will probably have zero actual effect on the problem of pedestrian safety. Maybe it will lead people to realize that we can do some infrastructure improvements or some systems approaches to improving safety. Yeah, but Kelsey, systems and infrastructure

improvements, as you've described, they take a long time, they can be contentious, they can be difficult, They can be expensive. And at the end of the day, I say, hey, you know, millions and millions and millions and millions and millions of people drive and need to drive all the time. And if six or seven thousand people are getting run over and killed in the course of a whole year in a country of 350 million, it's probably just not worth it. How do you feel about that argument? In my darkest moments, I feel like that too. But I think the best way to create a political constituency for change is to focus on two groups. One is the parents of children and teenagers in particular. We want our teenagers to be able to get to things. We want them to be able to go and play sports, engage in activities. Wouldn't it be

great to live in a world in which your teenager was able to safely get to the places they need to get to? I think the other approach is just to try to remember what it is that we've lost by driving everywhere. There is like the knowing of your neighbors, the moving of your body, the being a human in a place that is lost. And it's so hard to remember what it used to be like before we drove really and truly everywhere. So I'm hopeful that in the future, it will not just be a reduction in speeds and adding lighting, but it will include things like local breweries and local cafes and places that we can gather as neighbors in smaller communities.

But maybe that's too optimistic. What do you think? Is that too optimistic? There was a time when it was much more dangerous to fly in an airplane, but commercial airline crashes today are unbelievably rare. It took a lot of effort,

a lot of coordination and cooperation and regulation and training and technology to make that happen. What would it take to make auto travel that much safer? I would love to hear your thoughts. Our email is radio at Freakonomics.com. You can also leave a review on your podcast app. Coming up next time on the show. Please don't laugh, but we have made

a three-part series on the economics of whaling. Seriously. These are majestic creatures. But as long as our economic welfare depended on whales, no one seemed to take that attitude. I promise you, it is a fascinating story. It was an extremely lucrative and important industry. It's a surprisingly modern story, too, about labor economics. The more diverse a crew is, the more of a sense of shared purpose there might be. About new technologies. These things have grenades on them, and they're meant to hit the whale,

explode, and cause instant death. It's also about the evolution of the whale, from energy source to symbol of the environmental movement. But is the environmental movement also killing whales? It's definitely a concern that we've had all these whales washing up on shores. Everything you never knew about whaling. You can thank us later. That's next time on the show.

Until then, take care of yourself. And if you can, someone else too. You can start by not running anybody over. Freakonomics Radio is produced by Stitcher and Renbud Radio. You can find our entire archive on any podcast app or at Freakonomics.com, where we also publish transcripts and show notes. This episode was produced by Alina Kullman and mixed by Greg Rippin with help from Jeremy Johnston. Thanks for listening. I'd be willing to wager a bet that the people most annoyed by the delivery drivers in New York City are the people who have ordered Grubhub the most often. The Freakonomics Radio Network.

The hidden side of everything.