Q&A: Traffic Expert Says Using Tolls To Battle Congestion Is Mostly an Urban Strategy -- Though the Shinnecock Canal Presents an Opportunity

Not since the heady days of the original “Cops and Cones” program during the 2004 U.S. Open at Shinnecock Hills Golf Club has there been more experimentation on the South Fork’s highway system than this year.

New Southampton Town Highway Superintendent Charles McArdle has made it a mission to be aggressive in managing the morning and afternoon traffic jams on County Road 39 and Montauk Highway. He’s had the support of various local officials in trying to use cones, police officers and other resources to manage traffic flow.

On September 1, Town Supervisor Jay Schneiderman floated an idea at a Town Board session: What about using the bottleneck over the Shinnecock Canal — just three lanes of travel in each direction, east-west — to create tolls? Could charging people to use the roads during the busiest times, or to charge out-of-town contractors in particular for the privilege of being on local roads and adding to congestion, be a way to combat it?

“To what degree do we burden our local taxpayers for helping people from out of town get to work faster?” he queried, referring to ways to pay for the new traffic control efforts. He added, “I know it’s controversial to talk about, but that type of technology exists.”

In fact, tolls charged by E-Z Pass or license plate readers can be used to deploy congestion tolls, something being tried in some urban centers.

Dr. Michael Manville — who has never been to the East End — is a traffic expert, and an associate professor of urban planning at the UCLA Luskin School of Public Affairs in Los Angeles, California. His research and teaching focus, in part, on land use and traffic congestion. He agreed to field questions about the idea, in general, and how it might apply here.

How do tolls fit into a larger picture of how a community manages traffic? And are they only really good for really urban or really rural areas? Or is it something that could translate to an area like ours, which is more suburban/resort?

I hesitate to say it will work in an area like yours because I’ve never been there.

Understood.

But let me answer the broader parts of your question. A congestion toll, which is a very particular kind of toll, is usually not appropriate for a rural area, because rural areas, in general, don’t have much congestion. Congestion tolls are rare overall. They’re most common in urban areas, simply because that’s where more congestion is.

But, at least in principle — and we can talk a little bit more about the specifics of a particular place in practice — in principle, a congestion price can help anytime you have a congested facility.

Because the way people in my business understand traffic congestion is that, basically, the road is a valuable piece of land. The government has, by default, set its price at zero. And so what you have is a classic problem of an under-priced good. The demand exceeds the supply, and you have a shortage. The shortage leads to a queue.

The queue, in this case, is the long, slowly moving line of cars that we call traffic congestion.

And so if you set the price at a level that, “clears the market,” you’ll actually get less congestion and probably more throughput, right? Because congestion actually prevents the road from working well. It prevents the road from carrying more people.

There’s sort of blunt versions of congestion pricing in the world, like in central London. There’s very sophisticated versions of it on some toll roads in California. And in the central area of Singapore, it has, in fact, demonstrated an ability to reduce congestion, increase speeds, increase volumes and so forth.

Is there a principle about the use of the revenue generated? Is the revenue from that, should it be turned back into traffic-related spending? Or do other communities use that money for other uses?

So, this is an important point, because I think it’s an area where a lot of your readers might get confused, where a lot of people get confused.

The typical toll road in the United States is designed to raise revenue. You get on the New Jersey Turnpike and they’re collecting money from you, a bunch of toll booths, and that money is earmarked. It’s initially to cover the cost of the road, and now it sort of goes to various turnpike-related things.

But that toll is different from a congestion charge. The point of a congestion charge is not to raise revenue — it’s to reduce congestion.

And so, this is, like, it’s going to sound weird when I say it: In principle, you could take all the money raised by a congestion charge and sink it off Long Island — and it would still be highly beneficial, because you had made the road system work better.

Now, in practice, that’s a crazy idea. It’s money. People like money. Don’t dump it in the ocean. And so I think what ends up happening in many instances is that the revenue is allocated, the revenue becomes a political thing. It’s promised to different groups so that you can build support for congestion pricing.

New York City, down the road from you, is a great example of this. They want approval for congestion pricing in Manhattan, basically, by promising to use the money on the subway. But there’s nothing in sort of the transportation economics that says that that’s what you have to use it on. It’s just the politically astute thing to do.

If you’re looking at a community that might be considering using tolls as a way of battling congestion, are there certain elements that have to be in place to make it work? Certain geographic characteristics that you need to have, something like that? When you look at it, when does it work better?

So, one way it works better is just if the road that you are placing a toll on has relatively limited entering and egress points, and arguably relatively fewer road-based substitutes. And so to make this intuitive, placing congestion tolls on a freeway is much easier than placing congestion tools in a central business district, because there’s only a certain number of entry and exit points on the freeway. You just set up toll gate entries there, and you’re confident that no one else can get on the road. Whereas doing Lower Manhattan is going to involve a lot of different toll points, because there’s so many different streets.

The other nice thing about a freeway is that, yes, people could conceivably get off and drive on some nearby surface street, but the freeways are pretty unique in what they offer. And so you don’t have to worry as much about spillover.

And I think the third point is that it’s easier to do on roads when it’s clearer that the main goal of that road is to move a lot of vehicles at a reasonably high speed. So New York is going to put congestion tolls on the surface streets of Lower Manhattan, and I don’t think that’s a bad idea at all, but they’re going to have to balance their desire to have cars move faster through those streets with the fact that those streets are also used by pedestrians and cyclists and parades and so on and so forth. And so higher speed isn’t sort of an unambiguously good thing in Manhattan.

Whereas, if you were to look at the 405 freeway in California, the only purpose of that road is to move cars. And so you really get a lot of benefits if you can clear up the congestion.

So, if I present to you, then, that in our community we have two roads. One of them is a limited-access road with a speed limit of 55, and the other is a two-lane road, but they both cross the Shinnecock Canal, and they are the only two roads that do so. In some ways that is an optimal condition, in the sense of there aren’t a lot of complications. All of the traffic, it’s moving traffic across the canal. That’s the purpose for both of those roads. They’re the only two roads doing it, and there’s not a whole lot of concern about interaction with pedestrians. Right?

So with the obvious caveat, again, that I’ve never been there, that sounds like relatively favorable conditions for tolling.

Understood. And I don’t want to turn this into you endorsing this, necessarily. I get that. But the technology is really improved in this front, too. And that has a significant effect on how well something like this can work.

Yeah, I mean I think for a long time the idea of alleviating congestion using a toll is very old. If you go back to 1776, in “The Wealth of Nations,” Adam Smith’s landmark book of economics, he talks about a congested bridge and how to deal with it by just, at busy times, having the person collect a toll. So it’s not a new idea.

But the technology was a very limiting factor for a long time. I mean, now it’s more electronic, but back in the 1980s and 1990s, driving the New Jersey Turnpike, the toll booths themselves caused congestion, right? Everybody had to slow down and pay and get change. And so even if you had wanted to convert all those tollbooths into congestion pricing back then, the benefits would’ve been limited, because the transaction, the time, and effort of transacting to pay the toll, would’ve slowed the traffic down.

Whereas now you have E-Z Passes that let you just maybe slow down a little bit. And that really changes things a lot. It also, the new technology, allows the pricing to be dynamic, to seamlessly change as the demand for the road changes, which is important.

I would think the technology allows you to craft a policy as far as who’s paying tolls and when, and all of that. But that’s really the devil in the details, isn’t it? Who has to pay the toll and who might be exempted from the toll?

In our case, you have a lot of different components to those rush hours. You have commuters who live close by and work to the east. You have work crews that are coming from businesses from the west, servicing houses to the east. You have part-time property owners who are coming back and forth from the city to houses. And you have to decide who has to pay the toll and who doesn’t. Or is that not something you do with congestion tolls?

It’s not considered good practice. The congestion is caused because the car takes up space. And once the car is using space, other people can’t use it, and you get the shortage of road. The car takes up space regardless of how long you’ve lived in a place, whether you’re a part-time resident, whether you have another residence somewhere else.

The localist impulse to sort of punish the outsider always blooms large. But it sort of misunderstands the problem, because, I don’t know your community that well, but I imagine there’s enough locals there to cause a traffic jam.

But, more the point, it may not be legal. These are public roads. And then, especially, if any aspect of this is a state route, the State of New York will have something to say about your town saying everybody except people in the town has to pay. And it’s just extremely difficult.

Once you start saying certain people are exempt, how you going to do it? You could say, “Well, they get a special transponder,” or a special license plate or something. Trust me when I say that will start getting passed around. Trust me, when I said there’s probably a secondary market in them pretty soon.

We have beach passes out here, and we’ve been through this with beach passes.

Yeah, exactly. And so, I think that if the way we usually deal with this in principle, right, because congestion pricing is still pretty rare, is to say, “Look, if you’re using the road, you are getting a benefit from that road. And also you are imposing a cost on someone else. Someone else can’t use it while you use it. So you pay.”

The only exception to that would be if this would be a significant economic burden on you. In which case there should be some sort … An exemption is still not the way to do it, but maybe some sort of income support program paid for from the toll revenue.

And the corollary here is just imagine another large piece of network infrastructure like your electric grid, your water pipes, your heating fuel. What do we do with that, right, to prevent overuse and shortages? Everybody pays every time they use it. If you’re below a certain income threshold, you can participate in some sort of program that gets you a discount or a rebate or something like that.

And I know in our part of the world, electric rates are actually higher at peak demand times, for that very reason. And it’s the same principle, right?

Absolutely. It is not a coincidence that the only piece of network infrastructure in the country that has chronic shortages is the only network infrastructure that isn’t metered in some way. And that’s the roads.

While I have you, let me ask you this: Is adding more lanes a long-term solution to traffic issues in a community?

No. I mean, for the same reason that I mentioned earlier. The fundamental issue is that there’s a lot of demand and you’re giving it away for free. And so, if you add more lanes, what you’re going to do …

So, occasionally, someone will just say to you, “There’s no benefit at all to adding more lanes.” That’s not quite right.

If you add more lanes, what you’re going to do is that, for any given level of congestion, more cars will be sitting on the road. So say it takes 20 minutes to get from point A to point B on a stretch of road in Southampton. And then you have two lanes; at night, add a third one. What’s going to happen is that, because, again, lots of people want to get back and forth, travel times are going to speed up in the short run, because you’ve added an extra lane.

But then because travel time is sped up, people are drawn to the road — and the road fills up again. And so the equilibrium you arrive at in the mid to long term is just, it still takes 20 minutes to travel down that road, but instead of a hundred cars moving during that time, it’s 150. So that’s not nothing, but it’s not what most people think about when they think about “I want to reduce congestion.”

What most people think about is, “I’m sitting here, it takes me 20 minutes, I want it to take 15.” And adding a lane is not going to get you that.

Also adding a lane has real costs, it’s really expensive. Sometimes you have to take land, and that’s expensive. Or take someone’s house — that’s likely expensive and kind of terrible.

It means there’s more pollution during peak hours, and pollution at peak hours tends to be worse because the engines are moving slower, the vehicles are idling and so forth.

So the succinct way to say this is: If adding lanes was a long-term solution, Los Angeles and Houston would’ve solved the congestion problem, because there are places in our cities, Los Angeles and Houston, where the freeway is 12, 14, 16 lanes wide. And, every day, rush hour is full. And it’s ugly as hell.