Traffic: Why We Drive the Way We Do (and What It Says About Us) - Tom Vanderbilt (2008)

Chapter 6. Why More Roads Lead to More Traffic (and What to Do About It)

The Selfish Commuter

When a road is once built, it is a strange thing how it collects traffic.

—Robert Louis Stevenson

In the summer of 2002, a labor dispute at the ports of Los Angeles and Long Beach halted the flow of goods for ten days. Ships backed up, containers of Nikes and Toyotas lay dormant, and five-axle trucks, the kind that carry the containers from the ships to their destination, suddenly had nothing to haul. The impact on I-710, the route most trucks take from the ports, was immediate: In the first seven days of the shutdown, there were nine thousand fewer trucks on the highway.

Frank Quon, deputy district director of operations for Caltrans, the state highway authority, noticed something peculiar happening that week. The total traffic flow dropped by only five thousand vehicles. “Nine thousand trucks disappeared off the system,” Quon told me in his office in downtown Los Angeles. So why did the total flow drop by barely over half that? “Cars filled in the volume. Another four thousand cars just jumped in the mix.”

Almost instantly, drivers just seemed to know that the 710, where speeds jumped an average 67 percent during the shutdown, was a good place to be. They may have heard it on the traffic report, or a friend may have told them. Or they took it one day, learned that it was uncongested, and decided to take it the next day as well. What is curious is that the 710 was not necessarily sucking drivers off more crowded routes. “If you look at the parallel routes, like the 110 freeway,” Quon said, “the volumes remained essentially the same.”

It was as if drivers had suddenly materialized out of nowhere to take advantage of a highway that was, by Southern California standards, almost too good to be true. And it was: By the following week, when the ports reopened, the traffic was even worse than before the shutdown as trucks scrambled to catch up on deliveries—truck traffic, as you might have guessed, jumped much more than the total traffic. Now those new cars were deciding to stay away from the 710.

Engineers like Quon call what happened on the 710 a case of “latent demand.” “It’s the demand that’s there but because the system is so confined that demand doesn’t materialize,” Quon explained. “But when you create capacity, that latent demand comes back and fills it in.” Basically, people who would have never taken the 710 because it was too crowded suddenly got on. We don’t really know what they did before. Perhaps they used local streets. Perhaps they took public transportation. Perhaps they simply stayed home.

The point is that people are incredibly sensitive to changes in traffic conditions (sometimes too sensitive, as we shall soon see) and they seem capable of quickly adapting to even the most drastic changes in a road network. Engineers have a phrase: “It’ll be all right by Friday.” This rough rule of thumb means that even if on Monday something major happens that throws off the usual traffic patterns—a road is closed, a temporary detour set up—by the next Friday (or so) enough people should have reacted to the change in some way to bring the system back to something resembling normal. “When a change in a traffic pattern occurs, there’s a state of flux for a period of time,” Quon said. “We usually have everybody plan on expecting a two-week period. Things are going to keep balancing. Some days will be good, some days will be not so good, and then at the end of the two weeks, there will be an equilibrium in the system based on those changes.”

The latent demand that the newly fast 710 highway in Los Angeles had unlocked is often described by another phrase, “induced travel,” which is really just a twist on the same thing: There was a new incentive to drive on the highway. Imagine that instead of trucks disappearing from the 710, two new lanes were added. The result would be the same. Congestion would drop, but the highway would become more attractive to more people, and, when it was all said and done, traffic levels might be even higher than before. This is the “more roads create more traffic” argument you have no doubt heard before. It is actually an argument older than automobile traffic itself. In 1900, William Barclay Parsons, chief of New York City’s subway system, wrote, “For New York there is no such thing as a solution to the rapid transit problem. By the time the railway is completed, areas that are now given over to rocks and goats will be covered with houses and there will be created for each new line a special traffic of its own. The instant that this line is finished there will arise a demand for other lines.”

Over a century later, people are still arguing. There is a huge and enervating literature about this, which I heartily do not recommend. Do we build more roads because there are more people and more traffic, or does building those roads create a “special traffic all its own”? Actually, both of these things are true. What’s in dispute are political and social arguments: Where and how should we live and work, how should we all get around, who should pay for it (and how much), what effect does this have on our environment?

But studies suggest that induced travel is real: When more lane-miles of roads are built, more miles are driven, even more so than might be expected by “natural” increases in demand, like population growth. In other words, the new lanes may immediately bring relief to those who wanted to use the highway before, but they will also encourage those same people to use the highway more—they may make those “rational locators” move farther out, for example—and they will bring new drivers onto the highway, because they suddenly find it a better deal. Walter Kulash, an engineer at Glatting Jackson, argues that road building, compared to other government services, suffers disproportionately from this feedback loop. “You build more roads and you generate more use of the roads. If you add mightily to the sewer capacity, do people go to the bathroom more?”

If you do not believe that new roads bring new drivers, consider what happens when roads are taken away. Surely all the traffic must simply divert to other roads, no? In the short term, perhaps, but over time the total level of traffic actually drops. In a study of what they called “disappearing traffic,” a team of British researchers looked at a broad list of projects in England and elsewhere where roads had been taken away either for construction or by design. Predictably, traffic flows dropped at the affected area. Most of the time, though, the increase in traffic on alternative routes was nowhere near the traffic “lost” on the affected roads.

In the 1960s, as Jane Jacobs described in her classic book The Death and Life of Great American Cities, a small group of New Yorkers, including Jacobs herself, began a campaign to close the street cutting through Washington Square Park, in Greenwich Village. Parks were not great places for cars, they suggested. They also suggested not widening the nearby streets to accommodate the newly rerouted flow. The traffic people predicted mayhem. What happened was the reverse: Cars, having lost the best route through the park, decided to stop treating the neighborhood as a shortcut. Total car traffic dropped—and both the park and the neighborhood are doing just fine.

We have already seen how engineers’ models are unable to fully anticipate how humans will act on “safer” roads, and it is no different for congestion. It makes sense, mathematically, that if a city takes out a road in its traffic network, traffic on other streets will have to rise to make up for the lost capacity. If you removed one pipe in a plumbing system, the other pipes would have to pick up the slack. But people are a lot more complex than water, and the models fail to capture this complexity. The traffic may rise, as engineers predict, but that in itself may discourage drivers from entering a more difficult traffic stream.

Or it may not. Los Angeles currently operates with a freeway system largely built in the 1950s and 1960s. Its engineers never imagined the levels of traffic the city now sees. As John Fisher, head of the city’s DOT, put it, “They say, ‘If you build it, they will come.’ Because we didn’t build it doesn’t mean the people stopped coming. Freeways weren’t built, but the traffic is still coming anyway. There’s more and more traffic. The bottom line is that the L.A. area is going to be a magnet whether we build freeways or not. People are still going to want to come here.”

This raises the question of how much more successful a city Los Angeles could be if it had built all the freeways it never did, if one could magically whisk from downtown to Santa Monica in a few minutes. Then again, how desirable would a place like Beverly Hills be if the freeway that had been planned for it, to “cure” L.A. traffic, was now running through it? Wouldn’t the increased speed just attract even more people? Is traffic failing Los Angeles, or is it a symptom of a thriving Los Angeles? Brian Taylor, the planner at UCLA, argues that people often focus single-mindedly on congestion itself as an evil, which, leaving aside for a moment the vast, negative environmental impacts, misses the point: What great city has not been crowded? “If your firm needs access to post-production film editors or satellite-guidance engineers,” Taylor notes, “you will reach them more quickly via the crowded freeways of L.A. than via less crowded roads elsewhere.” Density, economists have argued, boosts productivity. Traffic engineers like to use the example of an empty restaurant versus a crowded restaurant: Wouldn’t you rather eat at the crowded one, even if it means waiting in line?

Users of Match.com, a dating service, are said, in places like Washington, D.C., to specify that they would like to meet someone who lives no more than ten miles away, presumably to avoid the hassles of congestion. Some have seen this as a social problem: Traffic is literally killing romance! Cupid is thwarted by congestion! This, too, misses the point: People move to places like Washington, D.C., in fact, because there are so many other people nearby. This is why cities play host to speed-dating events. There is so much “romantic congestion” packed into one room that daters must speed through all the potential choices. In Idaho, you will not face traffic trouble in driving well beyond the ten-mile range to meet dates; actually, you will probably have little choice. In any case, as anyone who has been in a long-distance relationship knows, those intervening miles can be a good way of deciding if a potential mate is really worth it.

What about all that time wasted in traffic? Surely that is costing us—$108 billion in the United States in 2000, according to one estimate. But a number of economists, most notably Anthony Downs of the Brookings Institution, have pointed out the potential flaws in these estimates. The first is that people seem willing to accept much of the delay, instead of paying to eliminate it (which means the “real” loss is closer to $12 billion). Another problem is that some models measure the costs of congestion against a hypothetical ideal of a major city in which all commuters could move at free-flowing speeds during rush hours—a situation that has not been possible since Juvenal’s Rome. Still another complication is that models judge the money people lose in traffic by a hypothetical wage rate, but this assumes that people would get paid for any time saved in traffic—or that they would somehow use the time saved in traveling to do something productive, not simply travel more. (As mentioned in the last chapter, many people seem to enjoy the time spent in their car.) Finally, no one really knows how much money we make because of our transportation system, so the losses due to congestion may be marginal. A useful comparison is the Internet. It imposes all kinds of costs on our productivity—YouTube videos, spam, fantasy football—but does anyone not think these are an acceptable cost for all the good we derive from it?

There is another way, a bit more subtle and complicated, that new roads can cause more traffic: the Braess paradox. This sounds like a good Robert Ludlum novel, but it actually comes from a classic 1968 paper by a German mathematician, Dietrich Braess. Put simply, the paradox he discovered says that adding a new road to a transportation network, rather than making things better, may actually slow things down for all its users (even if, unlike in the “latent demand” example, no new drivers have been induced onto the roads). Braess was actually tapping into the wisdom of a long line of people who had in some way thought about this problem, from the famous early-twentieth-century British economist Arthur Cecil Pigou to operations researchers in the 1950s like J. G. Wardrop.

You would need an advanced math degree to fully understand Braess and his ilk, but you can grasp the basic problem they were all getting at by thinking in simple traffic terms. First, imagine there are two roads running from one city to another. There is Sure Thing Street, a two-lane local street that always takes an hour. Then there is Take a Chance Highway, where the trip can be half an hour if it’s not crowded, but otherwise also takes an hour. Since most people feel lucky, they get on Take a Chance Highway—and end up spending an hour. From the point of view of the individual driver, this behavior makes sense. After all, if the driver gets off the highway and goes to Sure Thing Street, he or she will not save time. The driver will save time only if others get off the highway—but why should they?

The drivers are locked into what is called a Nash equilibrium, a strategic concept from the annals of Cold War thinking. Popularized by the Nobel mathematician John Nash, it describes a state in which no one player of an experimental game can make himself better off by his own action alone. If you cannot improve your situation, why move to a different road? The irony is that when everyone does what is best for him- or herself, they’re not doing what is best for everyone. On the other hand, if a traffic cop stood at the junction of the two roads and directed half the drivers to Sure Thing Street and half to Take a Chance Highway, the drivers on Sure Thing Street would get home no sooner, but the highway drivers would get home twice as fast. Overall, the total travel time would drop.

If all this puzzles you, Braess’s finding really makes the head spin. To simplify greatly, imagine again the two hypothetical roads I mentioned, but this time imagine that halfway between the two cities, Take a Chance Highway (where the trip takes less than an hour by however many fewer drivers choose it) becomes like Sure Thing Street (always an hour), and vice versa. Since each two-part route is likely to take the same amount of time, drivers split between the two routes, putting us in one-hour equilibrium.

But now imagine that a bridge is built connecting the two roads, right at the halfway point where Take a Chance becomes Sure Thing, and vice versa. Now drivers who began on Take a Chance Highway and found that it was not so good take the bridge to the other Take a Chance Highway segment. Meanwhile, drivers who began on Sure Thing Street are not about to cross the bridge and move to the other Sure Thing Street when, instead, they could stick around as their road becomes Take a Chance Highway (who knows, they might get lucky).

The problem is that if everyone tries to do what they think is the best thing for themselves, the actual travel time for all drivers goes up! The new link, designed to reduce congestion, has made things worse. The reason lies in what computer scientist Tim Roughgarden has called “selfish routing.” The way each person is moving through the network seems best to them (“user optimal”), but everyone’s total behavior may be the least efficient for the traffic network (“system optimal”).

This really brings us to the heart of traffic congestion. We are “selfish commuters” driving in a noncooperative network. When people drive to work in the morning, they do not pause to consider which route they could take to work, or at which time to take that route, so that their decision would be best for everyone else. They get on the same roads and wish that not so many others had also chosen to do the same thing.

As drivers, we are constantly creating what economists call, in the thorny language of economics, “uninternalized externalities.” This means that you are not feeling the pain you are causing others. Two legal scholars at the University for California at Berkeley have estimated, for example, that every time a new driver hits the road in California, the total insurance cost for everyone else goes up by more than $2,000. We do not pay for the various unsavory emissions our cars create—to take just one case, the unpaid cost of Los Angeles’ legendary haze is about 2.3 cents per mile. Nor do we pay for the noise we create, estimated by researchers at the University of California, Davis, to be between $5 billion and $10 billion per year. How can you estimate the cost of something like noise? Real estate provides a clue. Studies have shown that house prices decline measurably as traffic rates and speeds increase on the adjoining street, while, on the other hand, when traffic-calming projects are installed on streets, house prices often rise. One might argue that the lower price of a house on a high-traffic street already takes into account these costs, but what happens when a buyer purchases a house at a certain price and then traffic increases on that street, lowering its value? Living near a major road also exposes people to more hydrocarbons and particulates of car exhaust, and any number of studies have reported links between proximity to traffic and conditions like asthma and coronary problems.

There are other kinds of costs, more difficult to measure, that you as a driver put on the people you drive by. When the urban planner Donald Appleyard surveyed San Francisco in the 1970s, he found that on streets with more road traffic, people had fewer friends and spent less time outside. In the same way that traffic has been blamed for habitat fragmentation of the wild, cutting off species from foraging areas or reducing the tendency of birds to breed, high traffic helps starve social interaction on human streets (maybe this is how congestion hurts romance). Somewhat paradoxically, Appleyard found that people who lived on the streets with less traffic (who made more money and were more likely to own their homes) actually created more traffic themselves, while the people who lived on the high-traffic streets were less able to afford cars. The rich, in effect, were taxing the poor.

The most basic externality, however, is congestion itself. Your presence in the traffic stream helps add time to others’ commutes, just as others’ presences add time to yours. But no one driver is gaining more than those others are collectively losing. In economics, a “public good” is something that a person can consume without reducing someone else’s ability to consume that same thing or exclude them from doing so—sunlight, for example. An empty road late at night might be thought of as a public good, but a road with any kind of congestion on it quickly becomes “subtractable”—the more people who use it, the worse it performs.

This is the famous “tragedy of the commons,” as described by Garret Hardin, in which a pasture open to all is quickly filled up by herders who want to graze as many cattle as possible. Every time a herder adds a cow, he gains. The pasture eventually begins to suffer from overgrazing, but a herder still adds animals because he alone benefits from his gain, even if the returns are diminishing (and they ultimately vanish), while everyone shares the costs of that new animal. (Overfishing is another such oft-invoked “tragedy.”)

The “tragedy of the highway” is seen as every car joins the peak-hour freeway. As each car gets on, things get worse for everyone, but as there is still a gain for each driver (getting to work, getting home) that exceeds the gain from not driving, and as the loss is shared by all, people keep joining the freeway.

A Few Mickey Mouse Solutions to the Traffic Problem

98% OF U.S. COMMUTERS FAVOR
PUBLIC TRANSPORTATION FOR OTHERS

—headline in the Onion

So how can traffic congestion, this age-old dilemma, be solved? “Build more roads!” is a typical answer. “But more roads bring more traffic!” is the typical response. “Then build even more roads!” “But that will bring even more traffic!” Looking beyond that hall of mirrors, it’s worth pointing out a few things. The most obvious problem with building more roads to alleviate congestion is that we, in the United States at least, cannot afford them. Talk to just about any traffic engineer and they will repeat what the numbers already tell us: We do not have enough money to maintain the current roads, much less build new ones. What about all those fuel taxes? Drivers in the United States pay one-half the fuel taxes of drivers in Canada, one-fourth that of the Japanese, and one-tenth of the English. Adjusted for inflation, the fuel tax brings in less revenue than it did in the 1960s.

But even if we could afford to build more roads, that might not be the best way to spend the money. For one, as the transportation scholar Martin Wachs has pointed out, “Well over 90 percent of our roads are uncongested for well over 90 percent of the time.” Many congested roads are congested for only a few hours a day, which brings up the Wal-Mart parking lot problem of the previous section. Do you build a parking lot that will be below capacity for 364 days of the year so that it can accommodate every shopper on Christmas Eve? On the one hand, it might be a socially negative thing that some people have to get on the roads at five a.m. in Los Angeles to make it to work on time, or that both directions of the highway are crowded at many hours of the day. On the other hand, this is a good thing. It means the road network is being used efficiently. Empty roads may be fun to drive on, but they’re also wasteful.

Adding more lanes to a road is not always the traffic-busting silver bullet you might think it is. Imagine that you’re at the extremely crowded intersection of two three-lane roads. Why can’t they make it bigger? you ask. Look at all those people who want to turn left—why can’t they add another left-turn lane? The problem, as two Canadian researchers have pointed out, is that adding more lanes is a process of diminishing returns.

The bigger intersections grow, the less efficient they become. Adding a second left-turn lane, for instance, means that, for safety reasons, “permissive” (or on the green) left turns can no longer be allowed. Only “protected” left turns (on the green arrow) will be allowed. As fewer cars can now turn left on the green signal (through gaps in oncoming traffic), the arrow phase will have to be longer. This means most other movements have to be halted. More lanes also mean more “friction,” as engineers call it; a car wanting to turn left, for example, will find it harder going—and have a greater impact on the total traffic flow—when it has to cross three lanes instead of one. Given that bigger intersections take longer to cross, the clearance phase—that dead zone engineers introduce to make sure everyone has gotten through, including pedestrians—needs to become longer as well, further increasing delay. The result is that where an intersection with a single-lane approach would handle an average of 625 vehicles per hour, the next lane allows only 483 vehicles per hour, the third 463, and the fourth just 385. The more you spend on new lanes, the smaller the return—and the faster it becomes recongested.

Another problem is that most traffic jams are what engineers call “nonrecurring congestion.” This means a highway that normally functions fine is congested, perhaps because of construction or weather but, most often, because of crashes. Rather than build more lanes, the best congestion solution here is for people to get in fewer crashes—which, as described in Chapter 3, would happen if drivers simply paid more attention to their driving.

The actual crash, which may or may not close a lane, is only part of the problem, of course. The highway’s capacity drops an estimated 12.7 percent because of the line that forms—often on both sides of the highway—to take a look. This is where human psychology fails us. Not only do we have a morbid curiosity to rubberneck, but we feel we should not miss out on what others have had a chance to see. The economist Thomas Schelling points out that when each driver slows to look at an accident scene for ten seconds, it does not seem egregious because they have already waited ten minutes. But that ten minutes arose from everyone else’s ten seconds. Because no individual suffers from the losses he inflicts on others, everyone is slowed. “It is a bad bargain,” concludes Schelling. The ubiquity of cell phone cameras is making things worse, as “digi-neckers” slow things even more to take photos of incidents. To top it off, drivers looking at crashes quite often get into crashes themselves. A study by researchers at Virginia Commonwealth University found that the second-leading cause of distraction-related crashes (behind fatigue) was “looking at crashes, other roadside incidents, traffic, or other vehicles.”

What this means is that, at times, we have a perfect self-generating traffic jam: People slowing to look at crashes get into crashes, which causes other people to get into crashes, and so on. If traffic were a cooperative network and we could agree not to slow and look, Schelling notes, everyone could save time. Since that will never happen, traffic engineers have instead countered with antirubbernecking screens, which can be unfurled at crash scenes to block prying eyes. In theory these should help matters, but they have severe limitations. Just getting a screen to a crash site, past the traffic that has already developed, is hard enough. Then picture emergency responders, who probably have more pressing matters to attend to, trying to erect—in strong winds or snow—a giant wall of fabric, as if imitating the artist Christo. Plus, ironically, there is the interest in the screen itself. Janet Kennedy, a researcher at England’s Transport Research Laboratory, told me the screens had been tried on construction projects on the M25 motorway. “To start with it didn’t have much effect because people just looked at the screen anyway,” she said. “But already we’re finding people have stopped looking at the screen. They’re used to it.” That’s fine for construction sites, which the same people drive past each day. Unfortunately, this suggests that for crashes, the events that generate the most rubbernecking, the screens are of little help—the crash would be cleared long before drivers became accustomed to seeing the same screen.

But what about the congestion that’s “recurring,” that happens on the same roads every day? If money was available, we could build more lanes. Only this still does not get us past the pasture problem: Create a bigger pasture, and people will bring even more cows. Traffic congestion is a kind of two-way trap. Because driving is a bargain (drivers are not picking up the full tab for the consequences of their driving), it attracts many people to roads that are not fully funded; this not only makes them crowded, it makes it hard to find revenue to build new ones.

When Costco discounts televisions during its Christmas shopping promotions, pricing them so low that stores do not make a profit, what happens? There are huge lines at the door at five a.m. When cities provide roads that are priced so low that they lose money on them, what happens? There are huge lines on the highway at five a.m. Pricing changes behavior. This is hardly a revelation, but it’s always striking to see it in action. At a Pizza Hut in Beijing, I watched with some wonder as patrons at the salad bar carefully arranged towering piles of salad on their plates, then carefully walked away with mounds of teetering greens. Why did they do this? There was a flat fee for one visit, so patrons made sure they got their money’s worth. They traveled as efficiently as they could. What if the fee was good for unlimited visits to the salad bar? People would have made multiple trips, carrying smaller portions of salad. The traffic flow back and forth to the bar would have gone up.

In traffic, the basic model has been a state-subsidized, all-you-can-eat salad bar. Take as many trips on the roads as you like, whenever you want, for whatever reason. It may be a good deal for society—a loss leader, like Costco’s cheap televisions—but it’s such a good deal that everyone does it. Recently, however, as we have been running out of money and space for new roads, the thinking has turned from “How can we get more people on the roads?” to “How can we get fewer?” The answer, of course, is congestion pricing. As an idea, it’s hardly new. The idea of taxing people for the “externalities,” like congestion, that they create goes all the way back to economists like Arthur Pigou, who talked about the problems road users create for other road users in his 1920 book, The Economics of Welfare.

Later, the Nobel Prize–winning economist William Vickrey led a long, lonely crusade to get people to accept the idea that urban roads are a scarce resource and should be priced accordingly. After all, as Vickrey pointed out in 1963, hotels charge more for in-season rooms, railways and airlines charge more for peak travel periods, and telephone companies charge more during the times when more people are likely to call—why should roads not cost more when more people want to use them? (Vickrey was a bit ahead of his time: Told in the early 1960s that there was no way to track where people drove, or how much they drove, Vickrey, the story goes, built a cheap radio transmitter and installed it in his car, displaying the results to friends.)

Congestion charging, in cities like London and Stockholm, has been shown to work because it forces people to make a decision about—and gives them a precise benchmark against which to measure—whether a given trip is “worth it.” We may have been paying before, in time—which hardly helps fund the roads—but the human mind handles time differently than money. We seem less sensitive to the value of time, even if, unlike money, time can never be regained. It is easier for people to rationalize its loss. The problem with the crowded highway is that everyone suffers the same loss of time, even if some people’s use of the highway might be worth more to them—to take an extreme example, think of a woman about to give birth on the way to the hospital, stuck in a traffic jam alongside someone who simply “needed to get out of the house.” They may each feel that their trip is valid, but is that really how a scarce resource should be distributed?

When people are forced, by means of how much it will cost them, to think about when, where, and how they are going places, interesting things begin to happen. You might assume that a rush-hour highway is filled with people driving to work who have no other way to get there—and no other time they can travel—but studies suggest that this is not the case. When researchers have exhaustively tracked the license plates of every car traveling on rush-hour highways and matched the results to other days, they have typically found that only about 50 percent are the same people each day. Sometimes people’s patterns emerge when you look deeper into what would seem to be random behavior. In what the English traffic researcher Richard Clegg calls the “See you next Wednesday effect,” research has found that when people use a rush hour on Wednesday of one week, they’re more likely to be on that same highway on the next Wednesday than on another day.

Not everyone is so rigid in their habits. In 2003, a group of drivers in Seattle were outfitted with electronic devices that would tell researchers where and when they had driven. Baseline data was collected on these people’s typical habits. Then the drivers were informed that they would be given a hypothetical cash account. They would automatically be charged more for driving in the most crowded places at the most crowded times. Matthew Kitchen, director of the Puget Sound Regional Council, the group that sponsored the program (called Traffic Choices), said he was struck by how differently people acted day to day even before they were charged tolls.

Once the tolls kicked in, things really began to change: People left sooner, took different routes, took buses, “collapsed” trips into shorter bundles. “The reality which is emerging is that I think people are very intelligent agents, working on their own behalf,” he said. “They understand the unique trade-off they face between time and money. The range of response is extremely broad. For instance, my willingness to pay to save ten minutes today might be very different than tomorrow.”

How much did the charging affect driving? The total “tours,” as they are called in transportation-planning lingo, dropped by 13 percent. That may not seem like much, but in the world of bottlenecks, small changes can have big effects (a 5 percent drop in traffic, it is said, can increase speeds by 50 percent, even if that only means going from 5 to 10 miles per hour). With traffic jams, Kitchen noted, “Once you start falling off the cliff, you fall pretty fast and pretty hard. That’s why between 5 and 10 percent less traffic restores what are really credible speeds on the network. You don’t have to hit people over the head with something that is punitive. You can achieve reasonable results with incentives that result in fairly modest behavioral response.”

By getting just some people to change their behavior, congestion pricing can help reverse a long-standing vicious cycle of traffic, one that removes the incentives to take public transportation. The more people who choose to drive to work, the worse the traffic. This raises the time buses must spend in traffic, which raises the cost for bus companies, who raise the fares for bus commuters—who are being penalized despite their own efforts to reduce total traffic. As the bus becomes less of a good deal, more people defect to cars, making things worse for the bus riders, who have even less incentive to ride the bus.

It doesn’t take much to set this avalanche in motion. The historian Philip Bagwell notes that in 1959, only 7 percent of the total traffic entering London was via private car. But if just 1 percent of the people taking public transportation shifted to cars, the percentage of car journeys would rise 12 percent, and the number of cars in the traffic stream would jump by 5 percent. Which is exactly what happened, and London soon had “traffic thrombosis.” Everything engineers did to ease the flow just seemed to make it worse.

Congestion pricing reverses the cycle. Driving becomes more expensive, so traffic is reduced. The fees raised by pricing go into buses, which benefit in time and in money from the reduced traffic. This makes buses cheaper, and thus more popular. Small changes in traffic levels make all kinds of other things possible. In London, a familiar lament was the decline of Trafalgar Square, the city’s symbolic heart, home to Nelson’s Column and countless demonstrations through the years. But on most days it merely seemed the elaborate centerpiece of a busy traffic circle, a noisy and noxious holding pen for pigeon-feeding tourists. Then came a plan to close the street between the square and the National Gallery, uniting the two entities into a grand civic space. This was deemed, from a traffic point of view, impossible. As Malcolm Murray-Clark, the director of London’s congestion-pricing program, told me over tea in his office, congestion pricing changed all that. By removing the “background levels” of traffic from London, as he put it, planners had the wiggle room to remove the Trafalgar road without catastrophic consequences. “Eighteen percent of the traffic through Trafalgar Square did not have a destination in central London,” he said. “It was just a through trip. Those were the first to go, if you like.”

Congestion pricing is really just another spin on making the system optimal, or, to put it another way, saving people from their own instincts: How do you persuade everyone not to go to the same place at the same time? Cities like London are, in effect, learning from Disneyland. That may seem like a stretch, but consider that Disney theme parks open each day to a flood of people, many wanting to first go on the most popular attractions. Cities “open” each day with people all wanting to go to the same “attractions” at once. Disney executives are as much in the traffic business as the entertainment business: moving people around, from ride to ride (and through the shops and restaurants), in the most efficient manner and with the least customer grumbling. They hire talented engineers, like Bruce Laval, to manage these flows and queues.

Laval, now retired, joined the company’s industrial engineering department in 1971. His master’s thesis was on traffic signal coordination, and his first task at Disney was to figure out a way to reduce the wait times on its popular monorail. “Management wanted to put together justification to buy a sixth monorail train,” he told me. “They figured they needed more capacity to move more people.” But Laval ran simulations that came to a counterintuitive solution: Disney could move people faster by removing a train, not adding one. The reason was that each train had a buffer zone, for safety, in front of it; as it neared another train, it slowed or stopped. Reducing the number of trains meant they all moved faster (one of those “slower is faster” effects that show up so often in networks).

Early on, Disney realized that as the park grew in popularity, managing the queues of people would prove difficult, particularly on the marquee attractions like Space Mountain. What could you do? Disney could take the approach of our traffic networks, which is simply to let an inefficient kind of equilibrium take hold. Let people wait, and if the line is too long, they may decide on their own not to get in line (or get on the highway), and thus be diverted to other rides (roads). The queue will regulate itself. You can also make the line not seem as long, through various psychological tricks (like posting longer wait times than are really the case or having the queue itself wind through mini-attractions). But that still means people are waiting in lines (i.e., in traffic) and not being as productive as they might be, rather than shopping and eating (i.e., working or spending time at home). Disney could, and sometimes did, add capacity to its rides. But that, too, had limitations. “It costs a lot of money to add capacity,” Laval said. “If you eliminate wait times during your peak days, you’re over capacity for the other ninety-five percent of the year. You don’t design a church for Easter Sunday.”

So Disney tried a form of congestion pricing. It issued ticket books in which the tickets’ values reflected the capacity of the rides. Popular rides like Space Mountain required E tickets, which were more expensive than A tickets, good for tamer attractions like the Horseless Carriage on Main Street. The idea was not only to prevent people from simply lining up for the top attractions but to spread people out across the park, avoiding traffic jams at places like Space Mountain. “One way of increasing capacity is rerouting demand,” Laval said. This was successful to a point, but the signals that prices send can work in different ways. At Disney World, Laval explained, where 80 percent of park entrants were first-time visitors (Disneyland has more repeat visitors), many of whom had no particular itinerary of which rides to go on first, the E tickets were like a big red flashing sign saying, “Ride me first.” Everyone wanted to get their money’s worth, so they immediately gravitated to the most expensive rides. The rides were not only expensive because they were popular, they were popular because they were expensive.

Phenomena like this shows up in traffic as well: The HOT lanes in Southern California charge more as more people enter them (in order to help keep them from becoming congested); yet sometimes people enter a tolled lane precisely because it is expensive—they think the toll must be so high because the untolled lanes are really jammed. (This sort of behavior subverts the normal economics principle of “price elasticity,” in which the number of users should drop as the toll goes up.)

Disney finally hit upon the ultimate solution in 1999, when it introduced the FastPass, the system that gives the customer a ticket telling them when to show up at the ride. What FastPass essentially does is exploit the idea that networks function both in space and in time. Rather than waiting in line, the user waits in a “virtual queue,” in time rather than space, and can in the meantime move on to other, less crowded rides (or buy stuff). People can take a chance on the stand-by line, or they can have an assured short wait if they can simply hold off until their assigned time. Obviously, FastPass could not literally work on the highway. Drivers do not want to pull up to a tollbooth and be told, “Come back at two-thirty p.m.” But in principle, congestion pricing works the same way, by redirecting demand on the network in time.

Traffic can be made to flow better by redirecting demand in space, of course, if traffic engineers know what the demand and available supply are on a network at any given time—and if they can find a way to get that information to drivers. In the past, this has been a necessarily crude process, hindered by delays in getting and sending the information, and the ability to see the network at once with all its interacting flows. Surely you have had the experience of listening in vain to a rapidly spoken traffic report, hoping against hope to get the details on the jam you’re sitting in (and by some law, you never can). And as we saw in Los Angeles, traffic information often comes too late for us to do anything about it, or is not even accurate.

Rather than surgical strikes at congestion, one can always try carpet bombing. Sam Schwartz (a.k.a. “Gridlock Sam”), New York City’s former traffic commissioner, claims that by declaring “gridlock alert” days, he could “knock fifty thousand or sixty thousand cars out of traffic” by plastering the airwaves with dire warnings. “The Heisenberg principle exists in traffic. If you look at it and announce and tell people about it, it has an effect.” When he wanted to reduce traffic on one parkway so construction crews could work on an overhead rail link, he rolled out more horror stories. “I was able to scare away forty percent of the vehicles from that corridor,” he says. “We measured it. I was amazed at how effective we were. Sometimes when you hear on the radio, they talk about how terrible the traffic is—it’s really me, like the Wizard of Oz behind the curtain.”

But human psychology has a way of rearing its complicated head. One problem is that you can never quite know how people will react. In one study, researchers assembled a panel of drivers who regularly commuted on U.S. 101 in California’s Silicon Valley. Following a multiple-vehicle crash that took over a half hour to clear, causing extensive delays, the researchers interviewed the commuters. They found that only half the drivers had heard about the incident, and that even the majority of those drivers simply headed to work at the normal time, the normal way. Many people simply seemed unconvinced that they could save any time by changing their plans.

We have all had these moments. Do I take the local streets when I see there is a crash ahead? Is it better to leave early on Sunday morning to go back to the city, or will everyone else have that same idea? Do I get in the right lane because it seems empty, or is there a reason no one else is in it? It boils down to how we make decisions when we do not have all the facts. We rely instead on heuristics, those little strategies and mental shortcuts we all have in our head: Well, this road is usually busy for only a few minutes, so I’ll stay on it. Or: I bet that since the radio called for snow there will not be many people at the mall. We use our experience; we make predictions.

This recalls the famous “El Farol problem,” sketched by the economist W. Brian Arthur, after a bar in Albuquerque, New Mexico. The hypothetical scenario imagines that one hundred people would like to go to the bar to listen to live music, but it seems too crowded if more than sixty show up. How does any one person decide whether or not to go? If they go one night and it’s too crowded, do they return the next night, on the thought that people will have been discouraged—or will others have precisely the same thought? Arthur found, in a simulation, that the mean attendance did indeed hover around sixty, but that the attendance numbers for each night continued to oscillate up and down, for the full one hundred weeks of the trial. Which means that one’s chances of going on the right night are essentially random, as people continue to try to adapt their behavior.

This kind of equilibrium problem happens frequently in traffic, even when people have some information. In 2006, for example, the Dan Ryan Expressway in Chicago was undergoing massive repairs. The first day that eight express lanes were closed, traffic moved surprisingly well. The recommended detours moved more slowly than the highway. This was reported on the news. You can guess what happened on Tuesday: More people flocked to the highway. We can surmise that the expressway’s traffic went down on Wednesday, though it may equally likely have gone up.

What happens when we no longer have to guess? We are now just at the beginning stages of a revolution in traffic, as navigation devices, increasingly often equipped with real-time traffic information, enter the market. The navigation part alone has important consequences for traffic. Studies have shown that drivers on unfamiliar roads are roughly 25 percent less efficient than they should be—that is, they are lost—and that their total mileage could be cut by 2 percent if they were always shown the best routes. Logistics software now helps cut delivery times and fuel emissions for UPS and other truck fleets simply by finding ways to avoid, when possible, time-consuming left turns in two-way traffic. But the biggest change will occur when each driver can always know which roads are crowded and what alternate routes would be best—not through guesses but through accurate real-time data.

In theory, this will help reduce inefficiency in the system. Drivers are told there has been a crash ahead, and their in-car device gives them another route that is estimated to save ten minutes. But nothing in traffic is ever so simple.

The first problem is that real-time data is not yet what its name promises. At Seattle’s Inrix, for example, one of the key providers of traffic information, traffic-pattern data is gathered from a variety of sources, current and historical—from loops to probes on commercial vehicles to the schedule of conventions in Las Vegas, some five billion “data points”—and weighted according to its perceived accuracy and age. “So a thirteen-minute-old traffic speed estimate from a Caltrans sensor in the Los Angeles market would get a sub-five-percent weighting in our estimate of current conditions,” explains Oliver Downs, Inrix’s principal research scientist. Inrix estimates the current conditions every minute, but as Downs notes, “it’s a 3.7-minute-old estimate of the current conditions.” Customers, meanwhile, get a new feed every five minutes. “When we say ‘real-time,’” Downs says, this means “less than five minutes.” That might not seem like much, but, as Downs admits, “it’s long relative to how quickly things can change on the roadway.”

The other problem comes in how people will use that information, or what you should tell them to do based on the information. Michael Schreckenberg, the German physicist known as the “jam professor,” has worked with officials in North Rhine–Westphalia in Germany to provide real-time information, as well as “predictive” traffic forecasts. Like Inrix, if less extensively, they have assembled some 360,000 “fundamental diagrams,” or precise statistical models of the flow behavior of highway sections. They have a good idea of what happens on not only a “normal” day but on all the strange variations: weeks when a holiday falls on Wednesday, the first day there is ice on the road (most people, he notes, will not have yet put on winter tires), the first day of daylight savings time, when a normally light morning trip may occur in the dark.

This kind of information, along with the data gathered from various loops and sensors, can be used to make precise forecasts about what traffic will be like not only on “normal” days but when crashes or incidents occur. There is, however, a problem: Does the forecast itself change the way people will behave, thus changing the forecast? As the economist Tim Harford notes about Wall Street forecasting, if everyone knew today that a stock was going to rise tomorrow, everyone would buy the stock today—thus making it so expensive it could no longer rise tomorrow.

Shreckenberg calls this the “self-destroying prognosis.” In his office at the University of Duisburg-Essen, he points to a highway map with its roads variously lit up in free-flowing green or clogged red. “The prognosis says that this road becomes worse in one hour,” he says. “Many people look at that and say, ‘Oh, don’t use the A3.’ Then they go somewhere else. The jam will not occur since everyone turned to another way. This is a problem.” These sorts of oscillations could happen with even short lags in information, in what Shreckenberg calls the “ping-pong effect.” Imagine there are two routes. Drivers are told that one is five minutes faster. Everyone shifts to that route. By the time the information is updated, the route that everyone got on is now five minutes slower. The other road now becomes faster, but it quickly succumbs to the same problem.

This raises a question: Has the information provided actually helped drivers or the system as a whole—or has it triggered the “selfish routing” mentioned before? Moshe Ben-Akiva, the director of MIT’s Intelligent Transportation Systems program, has studied such travel behavior issues for decades. He calls traffic predictions a “chicken-and-egg problem.” “The correct prediction must take into account how people are going to respond to the prediction,” he says. “You cannot predict what will happen tomorrow without taking into account how people are going to respond to the prediction once the prediction is broadcast.”

And so researchers create models that anticipate how people will behave, based on how they have behaved in the past. Shreckenberg, in Germany, wonders if this means, in essence, not giving drivers the whole picture. “You have to structure the information. What you want is for the people to do certain things. Telling them the whole truth is not the best way.” This is something on the minds of the big commercial providers of traffic information. As Howard Hayes, vice president of NAVTEQ, said at the firm’s headquarters in Chicago, “What happens if once this really good predictive traffic information becomes available, everyone starts getting shunted over to a different direction, which itself becomes jammed? Ideally you need something sophisticated, so that a certain number of people get shunted to one route and others to another.”

Since the information is still so limited, and since so few people actually have access to it, we do not really know how it will all play out once everyone is able to know the traffic conditions on every road in a network. Most simulations have shown that more drivers having more real-time information—the closer to actual real-time, the better—can reduce travel times and congestion. Even drivers without real-time information can benefit, it is argued, because better-informed drivers will exit crowded roads, thus making those roads less crowded for uninformed drivers stuck in traffic. But as you might expect, studies suggest that the benefit for any one driver with access to real-time information drops as more people have it. This is, in essence, the death of the shortcut. The more people know the best routes at all times, the less chance of there being some gloriously underutilized road. This is good for all drivers (i.e., the “system”) but less good, say, for the savvy taxi driver.

Real-time traffic and routing is most valuable, it has been suggested, during nonrecurring congestion. When a road that is normally not crowded is backed up because of a crash, it’s useful to know of better options. During recurring congestion, however, those peak-hour jams that result from too many people going to the same place at once, the advantage shrivels once the tipping point of congestion has been passed. (It is most effective right on the brink, when alternative routes are on the verge of drying up.) In a traffic system that is always congested, any good alternative routes will have already been discovered by other drivers.

Another shortcoming of real-time routing is due to a curious fact about urban road networks. As a group of researchers observed after studying traffic patterns and road networks in the twenty largest cities in Germany, roads follow what’s called a “power law”—in other words, a small minority of roads carry a huge majority of the traffic. In Dresden, for example, while 50 percent of the total road length carried hardly any traffic at all (0.2 percent), 80 percent of the total traffic ran on less than 10 percent of the roads. The reason is rather obvious: Most drivers tend to drive on the largest roads, because they are the fastest. Even though they may have slowed due to congestion, they are still fastest. Traffic engineers, having built the roads, are generally aware of this fact, and would rather have you stay on the road that was designed for heavy use, instead of engaging in widespread “rat runs” that play havoc with local roads.

Both the promise and the limits of real-time traffic and routing information were demonstrated to me one day as I drove on Interstate 95 in Connecticut, using real-time traffic information provided by TeleNav via a Motorola mobile phone. The phone had been cheerily giving directions, even offering an evolving estimated time of arrival. Suddenly, an alert sounded: Congestion ahead. I queried the system for the best alternate route. It quickly drew one up, then delivered the bad news: It would take longer than the route I was on. The road I was on, congested or not, was still the best.

Real-time traffic and routing information and congestion pricing are two sides of the same coin. One tells drivers how to avoid traffic congestion; the other impels drivers to avoid traffic congestion. When the roads are congested, real-time information does little good, except to tell drivers, like the people in line for Disney World’s Space Mountain, how long they can expect to wait. This alone may be enough of a social good. But real-time congestion information, provided by the very cars generating that congestion, promises something else. It can be used to calculate the exact demand for any stretch of road at any time. With congestion pricing, the traffic on the roads will finally be made to act like the traffic in things, with market prices reflecting and shaping the supply and demand.