Traffic

by Tom Vanderbilt

  What had led Velkey, clipboard in hand, to parking lots? Interestingly, it was an offshoot of his prime research interest: the foraging behavior of animals, particularly how animals develop certain strategies in the face of constrained resources such as food or territory. He was studying this at the University of Montana, where wildlife abounds. It turned out there was an interesting example right outside the psych department window: the crowded parking lot. The value of the resource was clear—a faculty member had recently spent a day in jail after keying the car of someone who had stolen his parking spot. (Here we must remember the old dictum about what keeps a university running smoothly: “Beer for the students, parking for the faculty, and football for the alumni.”)

  In this lot, Velkey saw two kinds of behavior emerge: an active and a passive search strategy. Some people would drive around the lot looking for a space, while others would sit at the head of a row and wait for someone to leave. In terms of the avian foraging models Velkey usually studied, the active searchers were like condors, soaring and looking for prey; the passive searchers, meanwhile, were like barn owls, perched and lying in wait.

  Most people were active searchers, spending about as much time looking as it would take them to drive to the next available lot, while the smaller group would wait for minutes on end for someone to exit. This group, Velkey noted, almost always got a spot in the lot, while others found one elsewhere. (In that study the “postacquisition” costs of walking from the car were not measured, so it is hard to say who came out ahead in terms of total time.) A set of “evolutionarily stable strategies” had taken hold: If everyone tried to be condors, they would all be endlessly circling; if everyone tried to be barn owls, they would all be hovering around the same spot. Depending on circumstances (e.g., whether classes were about to let out), one strategy or another might bring more “local” success than another, but, Velkey notes, eventually everyone gets a spot.

  The way humans hunt for parking and the way animals hunt for food are not as different as you might think. Many scientists believe that animals’ foraging habits can be explained by a model known as “optimal foraging”—animals seek to gather the most food with the least effort (thus leaving them with more time and energy to, say, reproduce). These strategies evolve in response to the myriad numbers of life-or-death decisions that are made in each successive generation: Does the hunter go after the easy, low-protein prey or the elusive, higher-protein prey? How long does one stay in a particular patch before moving on to a possibly more productive patch? Does one look for food in a group or on one’s own?

  For some optimal foraging in your own backyard, consider the bumble-bee and the foxglove. Bees, it turns out, begin looking for nectar in the flowers arrayed on the bottom of the spike, slowly working their way up. Why? Because foxgloves add new flowers in an upward progression, so that those at the top contain less nectar. Bees also know to skip flowers they have already visited, and when a new bee lands on a foxglove that has already been visited by another bee, the odds are it will leave immediately. The chances of finding any missed nectar, it seems, are not worth the effort of looking.

  Now, instead of bees, think of humans parking. The parkers in the Montana lot who followed the “perching” strategy had evolved a very specific optimal strategy: They knew that near the top of the hour, as classes emptied, spots would become available, but it was better to search for the exiting driver than the spot. New visitors to the lot, however, or visitors who arrived too late, would circle in vain before ultimately deciding not to expend any more of their energy in this “patch.”

  In our daily lives as parkers, we face these foraging questions. We must decide whether to act like condors or barn owls. And we’re sometimes on the other end: It is not difficult to feel unnervingly like dying prey in the eyes of a stalking buzzard when you come out of a crowded shopping mall during the holidays and suddenly find yourself tailed by a creeping car. Is it faster to tail drivers to their cars and wait for them to load their merchandise or to look for an open space? Do we pass up less valuable spaces (i.e, “prey”) for higher-value spaces that might be lurking around the corner? In some cases in the animal world, it is better to hunt for food in groups, but in other cases, going solo is the better option. You may have experienced this dilemma as you cruised the streets of a city (or the rows of a mall) looking for a parking spot, realizing with a sudden dread that the person ahead of you, taillights flashing hopefully in front of potential spaces (which turn out to house fire hydrants or compact cars), is doing exactly the same thing. It no longer makes sense to look in the same places, as the car ahead will consume the resource first—better to head elsewhere.

  But neither animals nor humans always follow optimal strategies. One reason is that not enough information might be available—an issue that the parking industry is trying to address with technology that alerts people, via real-time signage or through cars’ navigation systems, to available (paid) parking spaces. Another reason might be the cognitive illusions I have already mentioned. Urban planners have pointed out that people seem willing to walk about a half mile from a parking spot to a destination. But they seem more likely to do so when they’re walking in the massive parking lot to a sports stadium, for example, than on downtown streets. There is an interesting explanation for this: Studies by geographers have shown that people tend to overestimate distances on routes that are “segmented,” versus those where the destination is in sight. Thus a football stadium a half mile away in a big parking lot seems closer than a half-mile walk involving multiple turns in a city.

  The Nobel Prize–winning economist Herbert Simon has suggested, in a seminal theory he called “satisficing” (a mix of satisfying and suffice), that because it is so hard for humans to always behave in the optimal way, we tend to make choices that leave us not with the “best” result but a result that is “good enough.” To take the bell-curve parking patterns described earlier as an example, drivers may have entered the lot with a general goal of getting the “best” spot, that is, in the row closest to the entrance. Once they were in the row, however, the goal changed to getting the best spot in that row. This is good in that it helps them feel satisfied with the spot they acquire. But if their strategy to get the “best” spot left them worse off overall, it might not be so good. Simon called the human limitations in making decisions “bounded rationality.” In Velkey’s study, people who focused on finding the “best” parking spot, in terms of distance, failed to account for all the time they were losing while searching—and they didn’t get closer anyway. We do not know if they were happy or not with their spot. When Velkey tried to conduct interviews, he was unsuccessful. Ironically, many people said “they didn’t have time.”

  The ways in which we hunt for parking, whatever their biological basis, are one of those subtle, almost secret patterns of traffic. They matter more than you might think.

  Parking occupies a strangely marginal place in the whole traffic equation. Engineers focus their energy on traffic-flow models, not parking models. We do not get morning “parking reports” on the radio. We tend to think of traffic as cars in motion; parking spaces seem more like real estate (indeed, they can be priced as high as houses, as the sale of quarter-million-dollar spots in New York and Boston has shown). But the simple, if often overlooked, fact is that without parking there would be no traffic. Every car on the road needs a place where it can begin and end, and mostly just sit there: Cars spend about 95 percent of their time parked.

  Parking is the innocuous gateway drug to a full-blown traffic-abuse problem. One survey found that a third of cars entering lower Manhattan were headed to free or subsidized parking spots. If those spots were not free or subsidized, there would be fewer drivers during the morning rush hour. Ironically, near the Department of Transportation itself, the streets are filled with DOT vehicles bearing special parking permits. How much do they add to peak-hour congestion? (This brings to mind a headline from the satirical newspaper the Onion: URBAN
PLANNER SITS IN TRAFFIC OF HIS OWN MAKING.)

  When the city of Copenhagen was looking to reduce the number of cars entering the central city in favor of bicycles and other modes of transportation, it had a very crafty strategy, according to Steffen Rasmussen of the city’s Traffic and Planning Office: Get rid of parking, but without anyone noticing. From 1994 to 2005, Copenhagen cut parking spaces in the city center from 14,000 to 11,500, replacing the spaces with things like parks and bicycle lanes. Over that same time, not accidentally, bicycle traffic rose by some 40 percent—a third of people commuting to work now go by bike—and Copenhagen has become one of the few places in the world where one can read, in a report, a sentence that would seem like a comical misprint almost anywhere else: “Cycle traffic is now so extensive that congestion on certain cycle tracks has become a problem, as has cycle parking space.”

  What you may not realize, when you find yourself driving on a crowded city street, is that many of your fellow drivers on that crowded street are simply cruising for parking. The problem is not so much the lack of street parking but the plentiful abundance of free or underpriced parking. This finding has sparked the fiery crusade of Donald Shoup, a bearded, bow-tied, and bicycling economist at the University of California, Los Angeles, and the author of a seven-hundred-page, cult-sensation tome titled The High Cost of Free Parking.

  The mantra used by Shoup, and his growing legion of supporters (dubbed “Shoupistas”), is the “85 percent solution.” In other words, cities should set prices on parking meters at a level high enough so that an area’s spots are only 85 percent occupied at any time. The ideal price, says Shoup, is the “lowest price that will avoid shortages.” Spaces with no meters at all, in a city like New York, are total anathema to Shoup. “People who want to store their car shouldn’t store it on the most valuable land on the planet, for free,” he told me in his office at UCLA, where a vintage parking meter sits atop his desk. “Something that is free is very misallocated.” This is why people who want to see free Shakespeare in the Park performances in New York City have to begin waiting in line as early as the day before (or hire people to do it for them), why cafés that offer free Internet access soon find themselves having to limit the time patrons can spend at a table, and why it can be so hard to find a parking spot.

  The reason people cruise is simple: They’re hunting for a bargain. In most cities, there is a glaring gap between the cost of a metered parking spot and that of an off-street parking garage. Looking at twenty large U.S. cities, Shoup has found that, on average, garages cost five times more per hour than metered street spots. The reason garages can charge so much, of course, is that the streets charge so little. When free parking spaces are available, the discrepancy is even higher, particularly for a free spot that can be held for many hours. And so people are faced with a strong incentive to drive around looking for parking, rather than heading into the first available garage.

  On the individual level, this makes sense. The problem, as is so often the case in traffic, is that the collective result of everyone’s smart behavior begins to seem, on a larger scale, stupid. The amount of extra traffic congestion this collective parking search creates is shocking. When Shoup and his researchers tracked cars looking for parking near UCLA (they rode bikes, so other cars would not think they were looking for parking and throw off the results), they found that on an average day cars in one fifteen-block section drove some 3,600 miles—more than the width of the entire country—searching for a spot.

  When engineers have tried to figure out how many cars in traffic are looking for parking, the results have ranged from 8 percent to 74 percent. Average cruising times clock in at anywhere from three minutes to thirteen minutes. What’s so bad about three minutes? you might ask. As Shoup points out, small amounts can have big consequences. In a city where it takes three minutes to find street parking, and where each space turns over ten times per day, each of those spaces will generate thirty minutes of cruising per day. At 10 miles per hour, that means the average space generates five miles’ worth of driving per day, which works out to a yearly sum that would get you halfway across the United States—not to mention a heap of pollution.

  But it is not simply that cars are driving while looking for parking. They’re driving in specific ways. There is the inevitable slowing to check out a prospective spot, the stopping to study whether a spot is valid, the actual jockeying into the spot, or what Shoup calls “parking foreplay,” in which the person detects that a space is about to be vacated and stops to wait. This may seem a minor offense, but as I discussed earlier, one car stopped on a two-lane street creates a bottleneck that cuts traffic capacity in half.

  This is worsened further by the inevitable delays and gaps caused by drivers battling to merge before they reach the stalled car. One person’s small act is felt by many. The famed urbanist William H. Whyte once espied this phenomenon during a traffic study of Manhattan. In his “mind’s eye,” he observed, one particular street was always “jammed” with double-parked cars (a result of underpriced parking, in Shoup’s view). But when he actually counted the number of double-parkers, he was shocked to only find “one or two” at any time. “It seemed odd that so few could do so much,” he wrote. “But the number, we found, was not the critical factor. It was the amount of time a lane was out of action because of double parking. Just one vehicle per block was enough.”

  The more time one spends looking for parking, of course, the greater chance one has to get in a crash, which then creates even more congestion. Interestingly, parking itself, according to some studies, is responsible for almost one-fifth of all urban traffic collisions. While some engineers think curb parking should be done away with entirely for safety and traffic-flow reasons, others counter that the rows of parked cars actually make things safer for pedestrians, both as a physical barrier and a source of “friction,” like street trees, that can drop traffic speeds by an estimated 8 miles per hour.

  To return to the Wal-Mart study mentioned earlier, the massively capacious big-box lots might seem to have little to do with crowded city streets. But there is plenty of cruising in large, free lots. It is simply that the incentive to save money has been entirely replaced by the incentive to save distance (and, theoretically, time, even if that ends up not being the case). In fact, there is always parking at Wal-Marts, so much so that the company lets people in recreational vehicles treat it like a campground. As Shoup points out, at places like Wal-Mart, the planners who dictate what size the parking lot should design for “peak demand”—that is, Christmas Eve—thus guaranteeing that most of the year, the lot has an abundance of empty spaces. The estimated demand comes from the parking-generation models of traffic engineers, which are filled, Shoup notes, with strange irregularities, like the paradoxical fact that banks with drive-up windows are required to have more parking spaces than banks without drive-up windows.

  Shoup argues that there is a circular logic at work in parking-generation models, one similar to that found in other kinds of traffic models. The demand for parking is treated as a foregone conclusion: Planners measure the number of people parked at a typical free parking lot in a location without much public transportation. The new Wal-Mart is built and, lo and behold, it attracts lots of cars. As Shoup writes, “The parking demand at new land uses with free parking then confirms the prediction that all the required spaces are ‘needed.’” Planners seem to ignore the fact that they are helping to dictate demand by providing supply. There are lots of cars in lots because parking is free.

  As Shoup reminds us, though, Wal-Mart’s free parking, like the free curb parking in cities, is not really free; the term is an oxymoron. We pay for “free” parking all sorts of other ways—and not just as a surcharge on the goods we buy. Parking lots are not only the handmaidens of traffic congestion, they’re temperature-boosting heat islands, as well as festering urban and suburban floodplains whose rapid storm-water runoff dumps motor oil and carcinogenic toxins like polycyclic aromatic hydroca
rbons (from shiny black sealcoat) into the surrounding environment and overwhelmed sewer systems. They represent a depletion of energy and a shockingly inefficient use of land—in a study of one Indiana county, Bryan Pijanowski, a geographer at Purdue University, found that parking spaces outnumbered drivers by three to one. The whole parking equation is like a large-scale version of that person at the mall, circling to get a “better” spot to save time and energy, and not realizing how much time and energy they have wasted looking for a better spot.

  Traffic patterns are the desire lines of our everyday lives. They show us who we are and where we are going. Examined more closely, this movement, like all desires, is not always rational or efficient. Traffic is a great river of opportunity, but often, as with the poor choices made with parking policy, we’re just spinning our wheels. In the next chapter, we’ll look at some more ways to get unstuck.

  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.”