Should Traffic Lights Be Abolished?

Kevin BERESFORD: Yes, I’m a member of the Dull Men’s Club. 

Stephen DUBNER: Let me make sure I’m understanding you correctly. It’s called the Dull Men’s Club, yes? “Dull” as in boring?

BERESFORD: Yeah. Dull is the new black. It’s sexy being dull. Women like dull men. They know we’re not going to run off with Lady Gaga or something. 

DUBNER: Yeah understood. We’re very reliable. 

BERESFORD: Exactly.

That is Kevin Beresford. In addition to the Dull Men’s Club, he belongs to another group that might sound dull.

BERESFORD: I’m the president of the U.K. Roundabout Appreciation Society, also known as Lord of the Rings. That’s my official title. 

A roundabout, if you don’t know, is a small, circular traffic intersection, typically with one lane. There is no traffic light or stop sign — more likely, a yield sign at each entry and exit. The traffic, therefore, flows continuously — if slowly — around a center island. A roundabout is not the same as a rotary, which is also known as a traffic circle. Rotaries tend to be substantially larger than roundabouts, with higher speeds. Roundabouts are much more common in the U.K. than in the U.S., and you could think of them as a kinder, gentler version of the rotary. So, that’s nice. But still: a roundabout appreciation society?

BERESFORD: All this came about in the year 2003. I ran a small printer in the town of Redditch, in Worcestershire, England.  

Redditch is in the West Midlands, near Birmingham.

BERESFORD: We didn’t have a lot going for us in that town. We had three prisons, but no cinema. We did have a 24-hour Tesco, but that was about our limit. What we did have was a copious amount of roundabouts. So just for a giggle — I’ll be honest here, just for a giggle — we ran off a roundabout calendar to give to our customers. And I couldn’t believe the feedback I got. It sold all around the world.

That first calendar was called “The Roundabouts of Redditch.” It inspired further editions: “The Best of British Roundabouts,” “Roundabouts of the World.” So it was only natural that a Roundabout Appreciation Society would follow.

BERESFORD: Yeah, in the old days, we used to meet on a bi-monthly basis in a pub called The Black Tap. We would swap data and roundabouts that we’d come about, maybe from where you’ve been on holidays.

DUBNER: And I hope you’ll get back to all that once the pandemic recedes. But when you get together in the pub to discuss all things roundabout — no offense, Kevin, but what’s there to discuss? 

BERESFORD: Well, it’s the aesthetic quality of the roundabouts. I mean, I’ve seen fountains, statues, planes, boats, trains, pubs, churches. There’s even, up in Yorkshire, a working windmill that actually produces flour. Can you believe that? On a roundabout! And that’s the beauty of a roundabout. Anything can go on a roundabout. 

DUBNER: Now, have you named your best roundabout of the year for 2020? 

BERESFORD: The English one is the Flanders Roundabout in Ashford, in Kent. It represents the First World War. You’ve got seven British Tommy soldiers, all with their heads bowed, with a vintage World War tank, and even the trees that have been planted there are from Flanders. It’s very poignant. But our international roundabout, now, that’s quite a quirky one. The international roundabout of the year is a gay gyratory. Can you believe it? It’s the first of its kind. It’s in Australia, in Canberra. 

DUBNER: And what makes it a gay roundabout? 

BERESFORD: It’s a rainbow, colored rainbow, roundabout. How beautiful is that? 

DUBNER: Oh, yeah, I see it now. I just looked it up. From above, it looks like a target, with the colors of the rainbow emanating out. It’s very beautiful. It’s very small, this one, yeah?

BERESFORD: Yeah. Size isn’t everything, is it? The fact is anything can go on a roundabout, and this is a great example. That city of Canberra, they overwhelmingly voted for the legality of same-sex marriages. So, that was the present from the gay community.

Kevin Beresford plainly appreciates the aesthetic possibilities of the roundabout. But roundabout aesthetics are not what we’re here to talk about today.

BERESFORD: If I came to America, and my mission was to put the roundabout into America, I’d save thousands of lives. In fact, that might be my mission, now, talking to you guys.

Could the humble roundabout really save thousands of lives? It is true that more than 35,000 Americans are killed each year in traffic crashes, and about a quarter of those deaths happen at intersections. All those crashes that kill 35,000 people also cause millions of injuries and nearly a quarter-trillion dollars in property damage, medical and legal costs, lost productivity, and more. What would all those numbers look like if some of our standard intersections with traffic lights were swapped out for roundabouts? Today on Freakonomics Radio: we consider this and many other roundabout questions, including the economics of the traffic intersection:

Jim BRAINARD: Couple of million dollars per intersection.

The environmental implications:

Doug HECOX: The emissions is less, significantly less. 

We consider a surprising technical complication:

Oliver CAMERON: You’d be stuck there at the entrance to a roundabout forever. 

And we wonder: why doesn’t America have more roundabouts?

*      *      *

DUBNER: Tell us a little bit about Carmel, Indiana. I know it is — I’m not sure if “suburb” is the right word — but it sits right next to Indianapolis. 

BRAINARD: We like to use the term “edge city.” Because it is a growing city on the edge of Indianapolis.

Jim Brainard is the mayor of Carmel, Indiana.

BRAINARD: Our city today is just over 100,000. When I became mayor in 1996, it had about 25,000 people in it.

DUBNER: Holy cow! 

BRAINARD: So we’ve had a lot of growth. It’s 50 square miles, which is double the size of the island of Manhattan. 

DUBNER: But a little less dense. 

BRAINARD: A little less dense. Although we’ve worked really hard to build a walkable, pedestrian-friendly downtown.

DUBNER: So as I understand it, Mr. Mayor, Carmel is particularly famous for — or at least prolific in its use of — a particular traffic pattern, yes?

BRAINARD: We have built more roundabouts than any other city in the United States or North America and perhaps the world. We have 133, with a handful under construction as we speak. There’s about 15 traffic lights left in our city. All but one of those I would like to see converted to a roundabout. Then we’ll pretty much be finished.

This one little edge city accounts for just under 2 percent of all the roundabouts in the U.S. The best estimate we could find puts the U.S. total at around 7,000. The U.K. has as many as 25,000 roundabouts, and of course it’s much smaller than the U.S.. France has as many as 50,000 roundabouts. So how did Carmel, Indiana, become an American outlier in roundabouts?

BRAINARD: I had seen a roundabout when I was doing graduate study in England, back in the early ’80s. And so I asked one of our consulting engineers to design a couple for a new road we were building on the east side of Carmel. 

This was in 1996, when Brainard became mayor.

BRAINARD: And he says, “No, I won’t put my professional stamp on them. They’re taking these things out of New England. They’re dangerous.” 

Brainard suspected that the engineer was confusing roundabouts with the much larger, higher-speed rotaries — a forgivable confusion, given the nomenclature. And rotaries were — and still are — quite common in New England. But at the time, Jim Brainard didn’t feel roundabout-conversant enough to challenge the engineer.

BRAINARD: And so the next weekend, I drove up to Purdue University, to their engineering library, and found a bunch of articles about the differences between roundabouts and rotaries. And I took these articles back and handed them to the engineer and said, “Please read these.”

Brainard is as pro-roundabout as a person can be — as evidenced by the 133-and-counting roundabouts in his relatively small city. I would put him somewhere between roundabout booster and roundabout evangelist. So I was curious to know if his boosterism is based on personal preferences, or something more empirical than that? In other words: what problems, exactly, does the roundabout solve? Let’s start with the most important problem: injury and death from vehicle crashes. In 2019, the most recent year for which we have complete data, crashes in the U.S. produced around 36,000 deaths and 2.75 million injuries.

As we noted earlier, about a quarter of all crash fatalities happen at intersections. So how do roundabout and non-roundabout intersections differ on fatalities? Looking at U.S. crash data from 2017 to 2019, you see that 0.1 percent of crashes at roundabouts result in a death. That could be the death of a driver, passenger, pedestrian, cyclist, anyone. 0.1-percent: that’s 1 death per 1,000 crashes at roundabouts. Okay, and how about your standard, four-way intersection, with traffic lights or stop signs? The death rate there is 0.4 percent, or 4 deaths per 1,000 crashes.

Even worse is what’s called a “Y” intersection — picture a capital “Y,” with a three-way convergence. For every 1,000 crashes at a “Y” intersection, there are nine deaths. Now, there could be confounding factors here; it could be that roundabouts tend to be put in areas where there’s less dangerous driving already, or areas where fewer people drive at night. But just looking at the data on fatal crashes, it would appear that roundabouts are much safer than other intersections. Why? Jim Brainard again:

BRAINARD: Roundabouts are smaller and because they’re smaller, everybody has to drive through them slowly. It’s about speed.  

Even compared to a rotary, roundabouts are slower: a large rotary allows cars to travel at around 40 miles per hour, with a lot of acceleration, braking, and ample opportunity for collision. The roundabout, meanwhile, forces vehicles to slow down to around 15 or 20 miles an hour.  At traffic-light intersections, meanwhile:

BRAINARD: Everybody has sped up, even when that stoplight is green, to get through before it changes. And sometimes we drive through a pink light too.

A pink light, if you don’t know, is what you call a traffic light that is just about to turn red.

BRAINARD: And the human error rate doesn’t really change. So the question is, what type of accident are we going to have? Well, if you sped up to go through a yellow light or a pink light or even a green light, and somebody makes a mistake —  it can be a very bad accident. At slower speeds, it’s not nearly as dangerous. And it’s safer for pedestrians, it’s safer for the handicapped, it’s safer for blind people.

But raw speed isn’t the only factor.

HECOX: Roundabouts allow the intersection to manage itself, in a manner of speaking. 

That’s Doug Hecox. He’s a spokesman for the Federal Highway Administration, or F.H.W.A.

HECOX: Drivers have to negotiate that circle a little bit more slowly, and it invites a little bit more of a conscientious driver by having to slow down and look around and accommodate oncoming traffic. And it’s just for a short period of time, usually just a few seconds, but that can be all the difference that it takes to save lives.

Hecox — and the F.H.W.A. — may not be as unabashedly pro-roundabout as Carmel mayor Jim Brainard, but they are pro-roundabout.

HECOX: There are significant impacts, pardon the pun, associated with having roundabouts.  

The F.H.W.A. helps U.S. states and territories manage 4.2 million miles of public road.

HECOX: They are a proven safety countermeasure, because the track record speaks for itself. The numbers tell the story. 

BRAINARD: So in the U.S., average fatalities per 100,000 per year is 14.

Jim Brainard again. That figure he’s citing — traffic deaths per year per 100,000 people — is actually a bit high. It’s now more like 12 deaths per 100,000 people. Still, how does that compare to Carmel?

BRAINARD: Carmel is at two per 100,000. 

DUBNER: And you’re going to attribute most of that benefit to roundabouts then, yes? 

BRAINARD: This is attributable to roundabouts. Because almost all fatalities happen in intersections because of the conflicting traffic flows. We had no fatalities last year in our intersections. The fatalities were in other places, a motorcycle that collided with a tree. An elderly gentleman that walked out right in front of a vehicle (we’re not sure why he did that.) 

We’ve done this chart of people in other cities around Indiana. One community has about 50,000 in southern Indiana, has 30 deaths per year per 100,000. Another one is in the 20s. Indianapolis is just below 12. And so you can then extrapolate their population of close to a million people,  and see if they had all roundabouts, how many people would still be alive that aren’t? It’s a very sort of sobering analysis when you do that.

There could of course be other, non-roundabout reasons why Carmel is so much safer. Maybe there’s less drinking and driving there. Maybe they have the best driver’s-ed. program ever invented. We should note that roundabouts may confuse some drivers.

BRAINARD: It tends to be how familiar the drivers are with that intersection.

Roundabouts do tend to produce more non-fatal crashes that lead to property damage, usually to the vehicles themselves. Driving on a curve rather than a straight line — especially if you’re unfamiliar with that particular intersection — that can lead to more fender-benders or encounters with the center island in the roundabout. Still, if your primary concern is the safety of human beings — the roundabout data are pretty persuasive. So you might expect that other places with a lot of roundabouts would also be safer — the U.K., for instance.

Again, there may be a lot of differences in driving between the U.K. and the U.S. But the fact is that the U.K. is swimming with roundabouts, and their overall rate of traffic fatalities is well less than half the U.S. rate. The U.K. traffic data also show that crashes at roundabouts are far less likely to be fatal than non-roundabout crashes, essentially mirroring the U.S. data. Now, let’s not pretend that roundabouts are a magic bullet, or the only bullet. Red-light cameras, for instance — as much as some people dislike them — they have also been shown to make intersections safer by cutting down on the number of drivers rushing to make the light. Red-light cameras do increase the incidence of rear-end crashes, as some drivers slam on their brakes to avoid a ticket. But these rear-end crashes are less dangerous than the “right-angle crashes” that happen when one vehicle blasts through an intersection that another vehicle is trying to cross at a perpendicular angle. So overall, the evidence in favor of the safety of roundabouts is robust. But wait. There’s more.

HECOX: Sure, sure. 

Doug Hecox, again, from the Federal Highway Administration.

HECOX: Many in the environmental community like the fact that because traffic isn’t stopped, like it is at a traditional signalized intersection, you don’t have vehicles idling and therefore the emissions from those idling vehicles is less. Significantly less. And so the air quality is improved. 

BRAINARD: Our city engineer has calculated how many tons of carbon we save every year. 

And Jim Brainard again, mayor of Carmel, Indiana.

BRAINARD: We think that we save on average around two, three million dollars a year of fuel for the general public by replacing stop lights.

Studies by transportation scholars have found that converting a standard intersection to a roundabout does significantly cut fuel consumption and carbon emissions. Transportation scholars point to yet another advantage of roundabouts: smoother traffic. Now, that might seem counterintuitive — at least it did to me when I first looked at this research. You’d think that the slow speed required by a roundabout — which is good for safety — would be bad for traffic flow. But the data say otherwise. The data say that roundabouts reduce congestion. Why is that? Well, think about how a traffic signal manages traffic.

Mike McBRIDE: A traffic signal is not efficient at all. 

That’s Mike McBride. He’s the former city engineer of Carmel, Indiana.

McBRIDE: That role can vary from city to city, but in the City of Carmel, the city engineer manages all transportation network issues. 

Including, in this case, the construction of roundabouts. During his tenure, McBride oversaw the building of roughly 90 roundabouts. As of the mid-1990’s, there were very few, if any, roundabouts in Indiana.

McBRIDE: But in 1996, Mayor Jim Brainard became the mayor of Carmel. 

At the time, McBride worked for an engineering firm, not the city itself.

McBRIDE: So the owner of the company came back to my desk and he said, “What do you know about roundabouts?” And I said, “What’s a roundabout?” So that was my introduction to roundabouts. He said, “Well, learn everything you can because you’re our new expert.” 

McBride did become an expert, and a convert. Especially when you compare a roundabout to an intersection with traffic signals.

MCBRIDE: Signalized intersections are definitely more familiar to people, so they’ve got a much wider public acceptance than roundabouts. But really, signalized intersections are designed for maximum efficiency, basically about an hour-and-a-half a day — maybe 45 minutes in the A.M. peak hour and 45 minutes in the P.M. peak hour. That’s about 20 percent of the daily traffic. We’ve all been sitting at an intersection, at a red light, when there are no opposing traffic cars where we’re just sitting there, burning fuel, wasting our time. Well, a roundabout, it does very well in those peak hours. And we know it’s safer. And at maximum efficiency is the other twenty-two-and-a-half hours a day, whatever that might be. So when you think about it from a sustainability and a reduction-of-fuel-consumption standpoint, roundabouts really have the upper hand.

Okay, if you’re like me, you’re starting to think that roundabouts sound too good to be true. They’re safer than other intersections. They lead to less fuel consumption and less pollution. They’re even better for traffic flow. So there must be a catch, right?

*      *      *

Let’s say you’ve been listening to this episode and you are a mayor or sit on a city council or run a transportation department, and you say to yourself, “Self, I think it’s time to get rid of some traffic lights and get me some roundabouts. They save lives and fuel; they cut down on pollution and congestion. Why on earth would I not want them?” Well, one thing we have not talked about yet is the cost of a roundabout versus a signalized intersection — especially if you want to convert an existing intersection. There are a lot of factors to consider here, including real estate: a roundabout can take up more space than a standard intersection. And if you’re in an older city, think how hard it could be to retrofit a roundabout onto an existing intersection. So let’s start with one of the most obvious costs of a traffic-light intersection: the traffic lights themselves. For that, we need a specialist:

Zachary CROCKETT: My name is Zachary Crockett, and I’m a reporter who’s obsessed with the economics of everyday things.

DUBNER: It seems like you cast your eye on a lot of things that a lot of people walk past and don’t think about. How do you do that? 

CROCKETT: One thing I do is I’m an active member of probably 200 Facebook groups. So, my feed is posts from Dogecoin traders and doomsday preppers, chicken farmers – all kinds of niche communities.

DUBNER: And we should say you’re not a chicken farmer or a Dogecoin investor. Is that true? 

CROCKETT: Neither of those is true.

One group that Crockett has been burrowing in with lately is traffic engineers — the people responsible for designing, sourcing, building, and maintaining signalized traffic intersections. The modern traffic light has been around since the 1920s. It’s hard to say exactly how many traffic lights there are in the U.S., but these engineers have a rule of thumb: there’s roughly one signalized intersection per 1,000 residents.

CROCKETT: If we hold that to be true, we’re looking at somewhere around 330,000 signalized intersections in the United States. 

And what does one intersection tend to cost?

CROCKETT: If you walk up to the standard intersection in San Mateo, California, where I’m from, you might see 16 different signal heads with three sets of lights each and then maybe four to eight poles. You’ve got four to eight of those push buttons for pedestrian crossings. You’ve got a bunch of underground wiring and brackets and custom-made hardware. So, that standard four-way signalized intersection, it likely costs more than your house. All in, you’re looking at anywhere from $250,000 to $1 million-plus.

Keep in mind Crockett lives in an expensive part of an expensive state, which may drive up some costs. In any case, let’s break down that pricey intersection.

CROCKETT: So, for starters, there’s the material costs and there’s the labor costs. And it’s about 50 percent for the design, the engineering, the development work, and then another 50 percent for the materials and construction.

On the materials side, let’s start with the signal heads — those rectangular boxes that contain the red, yellow, and green lights.

CROCKETT: Well, my favorite signal head is the Econolite, manufactured in the great city of Anaheim, California. They’re around $2,000 or $3,000 a pop. When they’re up there dangling in the sky, they look tiny. But when you’re confronted with one face-to-face, they’re behemoths.

And those behemoths need to be held up by something — typically, poles and supporting mast arms.

CROCKETT: There are really only a few companies in the United States that manufacture them. They can easily set a city back $25,000 a pop. And they can take up to 10 months to get. The reason they take so long is that really there’s some crazy engineering that goes into the process. They have to be hurricane-resistant. And these are thousand-pound beams that are hanging over our heads. There’s a lot riding on whether or not they stay in place.

Down on the ground, and essential to any intersection, is the control box.

CROCKETT: It’s like the size of a mini-fridge. And inside, there’s a bird’s nest of wires and flashing lights and computers. It’s kind of the brains of the operation. The controller usually runs around $30,000.

The electrification of an intersection can also be costly, depending on the circumstances.

CROCKETT: The wiring has a very dramatic range in cost. It can be like $3 a foot to $100 a foot. It’s not the wiring itself. It’s more what the wiring requires. Sometimes you have to do underground drilling and routing and you have to rearrange an entire intersection to accommodate for the wiring. 

And let’s not forget those pedestrian push buttons you see at the crosswalks — which, by the way, often don’t work.

CROCKETT: Altogether, those might run around $20,000 per signalized intersection.

And all this equipment will require maintenance and repair.

CROCKETT: Traffic-signal technicians — they’ve truly seen it all. The Florida hurricanes. Drunk drivers plowing into massive steel beams that require months of reconstruction. Rats for some reason are attracted to the sweetness of the wiring. Signal heads need to be constantly repainted and hardware needs to be re-affixed.

DUBNER: How much has the technology changed over the past, let’s say, 50 years for traffic lights generally?

CROCKETT: One of the major innovations in recent years, which is actually pretty simple, is the transition from incandescent to L.E.D. bulbs. So, a lot of cities have made that big switch. And it’s dramatically reduced their electricity bills. 

DUBNER: I know L.E.D. bulbs cost a lot more upfront, although they also last longer. Are there other downsides to the L.E.D. bulb? 

CROCKETT: So, one thing we’ve seen happen is that in Midwestern states like Illinois or Minnesota, the L.E.D. bulbs save on energy, but they actually didn’t produce enough heat to melt snow in the winter. So that obscures vision for drivers. And at one point, it caused a rash of traffic accidents. So, those cheaper L.E.D. bulbs had to be affixed with these special heat-lamp attachments that negated any savings that they would have had.

Crockett has also looked into what it costs drivers to sit at traffic lights, waiting for them to turn green.

CROCKETT: The Federal Highway Administration, they estimate that traffic signals account for about 295 million vehicle hours of traffic delays per year. If you work it out based on median household income figures, it’s about $2 billion in lost time.

So those are some of the costs of building and maintaining a standard traffic-light intersection. How does the roundabout compare? For that, we go back to Jim Brainard, the mayor of Carmel, Indiana, the roundabout capital of the United States.

BRAINARD: Well, there is a couple of different analyses we need to look at. First of all, if you’re converting a four-way stop to a roundabout, the roundabout is always going to be less expensive.

That is, if you’ve got an intersection with four stop signs, it will be cheaper to build a roundabout than it would be to add traffic lights, at least in Indiana. That’s because of all the traffic-light costs that Zachary Crockett told us about plus:

BRAINARD: So you got operating costs in addition to that. You have to send engineers out to reset the timing, which somehow gets off on a regular basis. And then every 25, 30 years, you have to replace that apparatus. And then you have a cost in a thunderstorm, the electricity goes out, you have to send a police officer out to direct traffic at a traffic light. The roundabout keeps working.

But what if an intersection already has traffic lights, and you want to convert it to a roundabout? Brainard says this easily pays off in the long run; but in the short run:

BRAINARD: There’s a substantial cost. Couple of million dollars per intersection, probably on average. Because you’re taking out that light, you’re probably buying some additional land in the corners. You have to move underground utilities out from under the light. 

DUBNER: And what’s that do to real estate and accessibility to that real estate — whether it’s shopping or doctor’s offices or so on. If I’ve got a four-way, traffic-light intersection, presumably you’ve got a business on every corner. With the roundabout, do you diminish your commercial access to the point where the cost savings from the roundabout itself, you’re maybe losing that savings because you’re surrendering commercial opportunity? 

BRAINARD: I don’t think so. You know, sometimes businesses say, “I like the cars stopped at a stoplight looking at my business.” Well, that’s not what we’re about. We’re trying to get traffic in and out safely and efficiently. We’re not meant to market your business with a stoplight. Here’s where it helps the businesses: you know, if you can’t get in and out of an area because it’s congested, people are going to avoid that area. We can move 50 percent more cars per hour through a roundabout than you can a stoplight. 

DUBNER: So here’s my big question. You make a very compelling argument for roundabouts. And you’ve built more roundabouts than, sounds like, just about anybody has. If they’re so great, why is every city in America not copying you? 

BRAINARD: A lot are. There is a large movement in the United States, to build roundabouts. There’s incentives in federal transportation law, particularly in areas that have bad air quality. Some state D.O.T.’s, Department of Transportations, are encouraging roundabout construction. I’ve been asked by a lot of cities across the country to tell our story, and try to help. But I think change is hard for humans. Change involves taking risks. When one takes a risk, you risk failure. It’s harder to try something brand new than to do what everyone else has always done. And, you know, politicians, elected officials are making these decisions. I think that’s been part of our hesitation. We have this vibrant, representative democracy, but sometimes we’re afraid to take risks because we want to get re-elected.

If you look all around the U.S., you do see a good bit of roundabout hesitation. In Flushing, Michigan, a roundabout project that had been under consideration for years was killed off because of fierce public opposition. In La Jolla, California, an existing roundabout was critiqued by one resident who wanted to add stop signs and speed bumps to make the roundabout more “civilized.” In Woodland, Washington, where the state Department of Transportation prefers roundabout construction, the city council voted to instead upgrade its signalized intersections. Why? As one council member said, “Everybody loses their minds and nobody knows how to drive in a roundabout.”

To be fair, people get used to what they’re used to; as Jim Brainard said, change is hard for a lot of us. But let’s be honest: one reason the roundabout remains unpopular in the U.S. is probably that it just seems too European. The city of Bath, in southwest England, is home to the iconic Bath Circus, or circle — three curved rows of townhouses that surround a ringed road with three vehicle entry points. It was completed in the 1760’s.  Some sources consider the first modern roundabout to be one in Görlitz, Germany, which dates to 1899. One person who subscribes to this “roundabout-is-too-European” theory is Kevin Beresford, president of the U.K. Roundabout Appreciation Society.

BERESFORD: It was that Lampoon with Chevy Chase — that’s so much stuck in the psychology of the Americans, that you’re just going to go ’round and ’round forever on a roundabout.

You may remember this scene:

Chevy CHASE: I guess what we do is just drive around the circle here.

 It’s from the 1985 film National Lampoon’s European Vacation. Chevy Chase plays the American dad Clark Griswold, driving around a London roundabout with his family.

CHASE: Hey look, kids! There’s Big Ben and there’s Parliament!

But the fears of La Jolla, California and Woodland, Washington are realized: he can’t get off the roundabout.

Ellen GRISWOLD: There it is! There it is! There it is!

CHASE: I know, I can’t seem to get over to the left honey, I’ll try next time. Sorry!

So he goes around and around as it grows dark outside.

Chevy CHASE: Kids!

KIDS: We know, Big Ben.

CHASE: Parliament.

HECOX: I’m not a psychologist, so I won’t pretend to give you a psychological analysis of the American driving public.

That, again, is Doug Hecox, with the Federal Highway Administration.

HECOX: But the Europeans have been using traffic circles in one form or another for far longer at a much higher percentage of the population than the United States has. 

The fact is that Columbus Circle, in New York City, was built in 1905. But for whatever reasons, the roundabout simply didn’t become popular in the U.S.

HECOX: The reality of road projects in general, whether they be roundabouts or big bridges, the response to and commitment to it by the public really does vary everywhere for any number of reasons. Some members of the public balk at cost issues. Others balk at the design of it. Some balk at the timeline of it. Others balk at the potential environmental downside of construction. And so it’s hard to pinpoint why there may be resistance to roundabouts, other than it’s an inconvenience or it’s going to force me to slow down. But we think they haven’t really understood just how beneficial roundabouts can be. Once the drivers use it and actually figure out how to do it properly, their apprehension tends to go away.

Indeed, one survey published in a transportation journal found that most drivers, just before a roundabout was built in their area, were anti-roundabout. Within six weeks of operation, about half the drivers approved. And once the roundabout was in place for at least a year, the vast majority of drivers approved.

McBRIDE: I think one of the other complicating factors is that in the United States we have the most established set of transportation design guidelines compared to other countries.

That, again, is the traffic engineer Mike McBride.

McBRIDE: That’s good for the most part. But what it does is it trains our driving population to rely on the transportation system to instruct them at every move. And it causes them to put a lot of trust, whether deserved or undeserved, in the technology that’s built into our transportation system. A roundabout is a deviation from that. Because now I have to think for myself. A yield sign at every entry at a roundabout requires that a driver has to come to that yield sign and think for themself: “Am I allowed to go? Am I not allowed to go?” So it’s uncomfortable for a driver that wants to be instructed.

CAMERON: A roundabout really does stretch the — how to phrase this — it’s almost like the processing power of the human brain.

That is Oliver Cameron.

CAMERON: I’m the co-founder and C.E.O. of Voyage.

Voyage is an autonomous-vehicle startup in California — think self-driving taxis and shuttle vans in places like senior-living communities. Big, spread-out areas that are often built, as it happens, with a lot of roundabouts. If our future includes more autonomous vehicles and more roundabouts, you have to wonder how well they’ll mix. When a human driver enters a roundabout, Cameron explains:

CAMERON: What you’re doing is having to do this intricate dance almost, with all of these other people. You’re basically taking social cues from all of these different vehicles and you’re doing that visually.

As we’ve been hearing, learning this roundabout dance can be a challenge for human drivers.

CAMERON: In a robo-taxi, that same challenge is present. You have to be able to perceive these objects early.

Objects including not just other vehicles but pedestrians, cyclists, dogs.

CAMERON: You have to be able to understand their intent, understand their direction, their speed. And all of these things make for a particularly interesting challenge. And a robo-taxi is not quite as aggressive as a human driver. So it’s very conservative.

 And in some circumstances:

CAMERON: You cannot be conservative. It just doesn’t work. You’d be stuck there at the entrance to a roundabout forever.

Or, maybe, inside the roundabout:                    

CHASE: Kids.

KIDS: We know, Big Ben.

CHASE: Parliament.

So how did Voyage deal with roundabouts?

CAMERON: The way we think about roundabouts is, it is a negotiation, and that negotiation starts by detecting these objects. What is then important is to extract almost metadata from each object. Like speed, like the prediction of where that object is going to be in 5 seconds, 10 seconds. And then what you’re doing is playing forward a whole bunch of different scenarios. 

The robo-taxi calculates what it would do in each scenario, and then chooses the option it deems the safest.   

CAMERON: These are algorithms we’ve been refining now for nearly four years. It’s able to handle that sort of negotiation really quite elegantly.

So Voyage’s vehicles can now navigate roundabouts. As can the Nexo, a self-driving car from Hyundai. Tesla says its self-driving cars will be able to handle roundabouts but, as C.E.O. Elon Musk said last August, it “will take maybe a year or so to get really good at roundabouts worldwide. The world,” he said, “has a zillion weird corner cases.” By “corner cases,” Musk means circumstances that arise outside the realm of normal driving. Everyone knows that strange things can happen when you’re driving — with other vehicles, pedestrians and cyclists, the weather. How would an autonomous vehicle handle an intersection where one of those L.E.D. traffic-light bulbs failed to melt the snow? Oliver Cameron, from Voyage, believes that autonomous vehicles will eventually learn to manage just about everything.

CAMERON: I think, yes, on a certain time horizon. And that’s because these vehicles don’t just learn from one situation. They learn from thousands of other situations they’re exposed to, every stress test under the sun.

There is another large, and less technical impediment, that driverless vehicles will need to overcome.

CAMERON: We’re deferring control to software, then it will cause anxiety. So it will be a challenge to get public acceptance. But I have confidence that people will see the benefits, both safety and convenience benefits. And that transition won’t be as painful as perhaps other transportation changes over time.

Maybe, maybe not. If you believe that autonomous vehicles will, in time, save millions of lives and trillions of dollars — as I and many other people do — then you’re eager for the technology, the governance, and the public perception to move forward fast. But if you look at the humble roundabout as one tiny example of a transportation change, you do wonder if maybe the public-perception hurdle just might be the hardest problem for autonomous vehicles. That said, remember the survey I mentioned about roundabout acceptance: most people, before they were familiar with them, were opposed; acceptance happened relatively fast; and after a year, they probably couldn’t even recall why they were opposed. That’s the way a lot of us are with change. We fight it until it’s inevitable; then we accept it and pretend we never had a problem with it in the first place. Will this happen with the roundabout in America? Here, again, is Kevin Beresford, Lord of the Rings, describing a roundabout that contains one of his very favorite central islands.

BERESFORD: In Birmingham, where I’m from really, originally, there’s an island called Spitfire Island and that’s got the three storming Spitfires shooting up into the sky. And that was outside the Jaguar plant, the car plant. And it means a lot to me, because my mother worked there during the war, producing these Spitfires. So, quite sentimental over that one. And it’s the fact that anything can go on these roundabouts. They’re like blank canvases for these artists and sculptors and gardeners. In America, I mean, the sky’s the limit, isn’t it? In Detroit, you could have a car on the roundabout or a Tamla Motown Records sculpture or something. Because a lot of English roundabouts reflect what’s going on in that area, or a local artist will put some sculpture or painting. It’s where your imagination wants to go. And it puzzles me why America doesn’t embrace this idea.

Maybe in time. Thanks to all the transportation geeks who geeked out with us today — Kevin Beresford, Jim Brainard, Doug Hecox, Mike McBride, Zachary Crockett, Oliver Cameron, and of course thanks to you for listening. Without you, we’d just be talking in our closets. If you want to hear an older episode of ours about car travel, and the dangers thereof, check out “The Perfect Crime,” Episode 165. It’s about how the laws protecting drivers mean that killing pedestrians often goes unpunished.

*      *      *

Freakonomics Radio is produced by Stitcher and Renbud Radio. This episode was produced by Mary Diduch. Our staff also includes Alison Craiglow, Greg Rippin, Mark McClusky, Matt Hickey, Zack Lapinski, and Emma Tyrrell; we had help this week from Jasmin Klinger. Our theme song is “Mr. Fortune,” by the Hitchhikers; the rest of the music was composed by Luis Guerra. You can subscribe to Freakonomics Radio on Apple Podcasts, Stitcher, or wherever you get your podcasts.