Transportation Anywhere, Anytime, and for Everybody
MORE THAN FORTY YEARS AGO, WHEN I WAS IN GRADUATE SCHOOL AT the University of Pennsylvania, my mentor and advisor, Vukan Vuchic, often compared the state of transit systems in the United States with those in European cities. I especially recall his vivid description of the subway system in Moscow—not just the efficiency of the trains, but the beauty of the stations, which featured chandeliers hanging from the ceiling as if they were lighting nineteenth-century ballrooms.
Even more vividly, I remember Professor Vuchic saying that, in Moscow, it was common to see formally dressed couples on the subway: women elegantly turned out in long dresses, men in tuxedos, on their way to the opera or the theater. “Can you imagine,” he exclaimed, “someone riding the New York City subway wearing a tuxedo?”
In 1969, when I first heard Professor Vuchic on the subject, he had a point. It was two decades after the subway fare was doubled from a nickel to a dime, beginning a vicious circle of regular fare hikes and worsening service, and more than five years before the financial crisis that would mark an even steeper decline in New York’s transit system—and in just about every aspect of life in New York. Subways and buses weren’t exactly the first choice for the city’s more affluent residents. In a very smart 2014 column in the New York Post, Nicole Gelinas reminded readers that as far back as 1953, Esther, the poorly paid magazine intern who is the heroine of Sylvia Plath’s novel The Bell Jar, took cabs everywhere in order to avoid the smelly, dirty, not-for-nice-girls subways, especially when dressed for an evening out. Thirty years later, the heroes of Jay McInerney’s Bright Lights, Big City were, if anything, even more repelled by New York City’s public transit. By then there was plenty of dressing up and revelry, but the subways are mentioned only a few times, and never positively: “At the subway station, you wait fifteen minutes for a train [until] a local, enervated by graffiti, shuffles into the station.”
But thirty years after that? In the spring of 2014, I left a black tie affair at the Waldorf Astoria on Park Avenue and 50th Street, walked four blocks to the subway station at 53rd Street, and was home in fifteen minutes. Me and my tuxedo and my wife dressed to the “nines.” On the subway.
The point of this story isn’t to advertise my virtue, or to give a shout-out to the investments made in my hometown’s public transit system over the last thirty years. It’s not even to demonstrate how a great public transit system—one that operates not just everywhere, and all the time, but for everyone—builds social cohesion. Not exactly, anyway. It’s this: because great public transportation systems are expensive, they only get fully funded when they’re used by both the well-to-do and the not-doing-so-well.
This is one of the sad but true aspects of transportation, one that they don’t teach in engineering school. No matter how well laid out the sidewalks and bike paths of a city’s active transportation network, no matter how cleverly designed its multimodal grid, no matter how easily its residents can get real-time interactive directions, if the city’s public transportation becomes a system only for the less well-off, it’s in trouble. It’s the same phenomenon that hamstrings public hospitals and public education: unless every socioeconomic group in a particular city feels invested in the system, it starves. As much as anything else, this fact explains why, despite all the well-documented problems with our dependence on private automobiles, road building continues to have first call on transportation budgets. Streets and highways really are used by everyone. Whether we’re talking eighteenth-century streets like Bedford Avenue, or Houston’s hypertrophied Katy Freeway, roads are just as likely to carry a brand new Mercedes-Benz as a ten-year-old Chevrolet Impala. Buses and streetcars, on the other hand, are the opposite of economically diverse. In the United States, 63 percent of the users of small transit systems, 51 percent of users of medium-size transit systems, and even 41 percent of riders in the largest transit systems are at or below the poverty line.
Even worse, the tendency of public transit systems to be perceived as the choice of travelers who can’t afford something better is vulnerable to what engineers call positive feedback: small pushes in one direction (either good or bad) tend to accelerate movement in that same direction. It can become a vicious circle: the more transit becomes dominated by less affluent people, the more it becomes associated with poverty. And the more it gets associated with poverty, the less appealing it becomes for the affluent. Equity declines.
You’d be forgiven for thinking that transportation had been at the front line in the struggle for equity in the United States for more than a century, though not always in the way that I’ve used it above. The term appeared prominently in the names of two vast multiyear federal transportation bills in the last two decades: the 1999 “Transportation Equity Act for the 21st Century” and its 2005 successor, the “Safe, Accountable, Flexible, Efficient, Transportation Equity Act [SAFETEA].”a In both cases equity is used to describe a fair, or at least not too unfair, allocation of federal highway funds among the states.
Those multiyear transportation bills are all about something often called return-to-source or horizontal equity: a bit of jargon that describes a point at which states, municipalities, and even individuals, in the words of Todd Litman of the Victoria Transport Policy Institute, “get what they pay for, and pay for what they get.” SAFETEA, for example, guaranteed that each state get back between 90 and 92 percent of its residents’ contribution to our old friend, the Highway Trust Fund.
Redistributive or vertical equity, on the other hand, has a different definition of fairness. This kind of equity recognizes that since some groups are advantaged, others must be disadvantaged, and, to balance the inequalities of the private sector, the disadvantaged should be favored in public transportation policies. Offering discounted fares to less affluent riders, or increasing bus routes in poor neighborhoods, for example, corrects for the fact that not everyone starts life in an affluent family. Investing in buses that can accommodate wheelchairs balances scales that are out of kilter in another way.
A lot of the equity discussions today, as above, are concerned with the competing demands of relatively well-off drivers and less affluent transit riders. But even within the world of public transit, scarce resources have to be allocated either horizontally or vertically. In fact, long before the automobile transformed travel, there were still pretty pointed debates about the allocation of public transit resources—usually between rich and poor, even more frequently between black and white. The battle for civil rights in America was famously fought out in streetcars, trams, buses, and trains.
In 1892, a black man named Homer Plessyb bought a ticket on the East Louisiana Railroad and took a seat in the “whites only” car, thus deliberately challenging the state’s Separate Cars Act. After Judge John Ferguson ruled that Plessy had to pay a fine for his presumption, the appeals that followed ended up in the US Supreme Court. When the Court handed down its decision in Plessy v. Ferguson, it upheld the constitutionality of “white” and “colored” sections, enshrining the concept of “separate but equal” facilities for whites and African Americans for the first half of the twentieth century. Five decades later, Rosa Parks refused to take a seat in the back of a public bus in Montgomery, Alabama, igniting the yearlong boycott that was ended by another Supreme Court decision, this time desegregating the city’s buses and consequently public transit throughout the United States.
Over and over again, access to public transportation and the promotion of social equality have been joined together at the hip. This isn’t just some vague Progressive liking for diversity for its own sake. Smart streets are diverse, but it’s not a cost: it’s a benefit. Neglecting this is one reason that the streets of so many planned communities, from Radburn, New Jersey, to Columbia, Maryland, aren’t as smart as their designers had hoped. Smart streets are more than just paths through well-designed theme parks, and they’re the opposite of exclusive. In order for a community to be vital—to be alive—its streets have to welcome the widest variety of people, precisely because that’s what makes the streets interesting and appealing in the first place. Transportation policies that segregate people by income or education aren’t just unfair, they’re self-defeating.
For most of American history, the challenge of fighting that kind of segregation was simple, though demanding: assuring access to the disenfranchised. Today, however, transportation planners have to balance two interests that aren’t always in sync. On the one hand, we’re obliged by every measure of decency to provide access to the people who need public transit the most, but on the other hand, we have to make it an appealing option for the people who need it far less.
The first objective is clear enough. Poor people generally need and use public transit far more frequently and intensively than anyone else. The nation’s poorest families spend more than 40 percent of their take-home pay on transportation, including cars. African American and Latino poor families spend most of that 40 percent in subway turnstiles and bus fare boxes.
The fact that less affluent families allocate their transportation budgets differently from the more well-to-do is mirrored in the ways that the state, municipal, and national governments allocate their budgets. This makes for a lot of racially tinged transit policies, all of them costly to poor people. Almost eighty cents of every federal transportation dollar already goes to highway building and road maintenance, which obviously discriminates against people who aren’t affluent enough to own cars. Even the remaining 20 percent—billions of dollars—spent on public transit isn’t distributed very equitably. Over the last thirty years, attorneys representing black and Latino plaintiffs have won dozens if not hundreds of lawsuits alleging discriminatory funding practices, in places as far apart as Los Angeles, Macon, Atlanta, Houston, and Boston (particularly the disproportionately black suburbs of Roxbury and Dorchester). Frequently, these transit policies robbed the bus systems used disproportionately by poorer, urban families in order to subsidize the train lines used by affluent suburbanites: Robin Hood, only backwards.
Different access to government funds isn’t the only form of bias faced by the less affluent (and less white) segments of society. Private businesses engage in different but no less costly forms of discrimination in their attempts to keep the wrong sort of people away. As a planner and engineer I’ve encountered all kinds of ways that retailers in shopping malls, for example, try to deter bus riders from visiting, up to and including lobbying municipal governments to prevent city buses from stopping anywhere conveniently close. Even when the shopping malls “allow” buses, they generally send them to the back of the mall parking area, behind the buildings. Buses that once relegated people of color to rear seats are now themselves forced to the back of parking lots.
When the Swedish furniture emporium IKEA planned to open a store in Brooklyn, I advised them to try something different from their traditional footprint, which was that of a very large store surrounded by an even larger parking lot. It was a precedent that had worked well for them, and seemed impossible to change, since their retailing model depended heavily on customers who could transport hundreds of pounds of flat-packed wooden furniture from store to home. However, I reminded them, they were trying to sell their goods in a market in which fewer than half the households owned cars. “How about giving the buses the preeminent spot, right in front of the store?” I suggested.
To my delight they agreed. Today, not only do two New York City bus lines stop right at the front of the store, so does an IKEA-financed free ferry to and from Manhattan, as do free buses from nearby subway stations. The result is a little discombobulating: a hugely successful branch of IKEA has a parking lot that is always half empty.c
The IKEA story, though, remains an unusual one. Discrimination against public transit is still endemic. Sometimes, it’s also deadly. In 1995, Cynthia Wiggins, who lived on the largely African American east side of Buffalo, New York, took the Niagara Frontier Transportation Authority’s Number Six bus to her job at the suburban Walden Galleria Mall. However, since the mall management had promised its tenants that no inner city bus would be allowed to stop at the upscale mall, Cynthia’s bus dropped her more than a quarter-mile away. On December 14, she was hit and killed by a dump truck while trying to cross a seven-lane highway lined with eight-foot-high barricades of snow.
Then there’s Atlanta.
Atlanta doesn’t, of course, get anything like Buffalo’s annual allotment of snow, which is probably a good thing. On Tuesday, January 28, 2014, a snowstorm that would have been laughed off in Buffalo—maybe two inches of powder and wet snow—paralyzed Georgia’s capital. Thousands of flights were canceled, and more than two thousand kids spent the night in schoolrooms and police stations, unable to get home. Two weeks later, the city hit the nation’s front pages again, when another storm, this one of ice, did it again, only worse. A city of six million, held hostage to an accident of weather. Or so it seemed.
However, Atlanta’s Snowpocalypse wasn’t the result of bad meteorological luck, but bad transportation policies, policies that were a predictable result of what a charitable person might call racial—or at least class-based—Balkanization. In the same way that a map of Vancouver reveals how geography can support a healthy transit network, a map of Atlanta shows how it can cripple one.
The big difference is that Vancouver’s transportation geography was determined by natural boundaries, but Atlanta’s was manufactured. The Atlanta metropolitan region comprises (depending on who’s doing the counting) as many as two dozen separate counties. Within those counties are more than sixty municipalities. That’s sixty mayors, or city managers, or city councils, all with slightly different ideas about how well they want to play with the other mayors, city managers, and city councils. The result is that the city of Atlanta became, in large part, a commuting destination for people who live in Atlanta’s suburbs, and a lot of them moved to those suburbs precisely in order to avoid paying for the schools, police, and—especially—the transportation network used by the people they had left behind. When they travel to Atlanta, on business and pleasure trips, most of them do so along one of the most congested parts of the entire Interstate Highway System, a so-called downtown connector made up of a confluence of I-75 and I-85, which not only did to neighborhoods like Mechanicsville and Summerhill what the Cross-Bronx Expressway did to East Tremont, but produces some of America’s worst traffic jams, even when the weather cooperates.
The downtown corridor is the overwhelming choice for Atlanta’s millions of commuters because it’s essentially the only one. While the Metropolitan Atlanta Regional Traffic Authority is the country’s eighth largest, it serves only the city of Atlanta itself, and two of the region’s counties, Fulton and DeKalb. Everywhere else, MARTA is forbidden to travel. (A clue why: a widely used version of the authority’s acronym is “Moving Africans Rapidly Through Atlanta.”) It wasn’t an accident of weather that made millions of people hostage to the most weather-sensitive form of transportation. It was a self-inflicted failure of understanding. Public transportation works only when it’s used by every segment of the public. When it’s equitable. When it’s transportation for everybody.
As traditionally understood, no transportation policy (much less any real-world transportation system) can both return benefits to each community commensurate with that community’s own contributions and restore fairness to disadvantaged communities. Most fail to do either. Resources are finite, the argument goes. Transportation everywhere, anytime, and for everybody is, like equality itself, a noble-sounding goal, but in the real world must give way to a series of compromises.
This is true as far as it goes. You can’t spend the same transportation dollar twice, which means that planners frequently have to limit service in one area in order to provide it in another. This is what the designers of Houston METRO’s “frequent network” did when they chose per-dollar patronage—the maximum number of riders the system could carry for a given amount of money—over providing an equal number of buses and bus stops to every part of the city.
As we saw back in Chapter 6, the frequent network is a very well thought-out response to a very difficult set of problems. But it isn’t the only route to transportation equity. Even if resources are finite, they aren’t fixed. When they can be increased, it makes it a whole lot easier to improve equity; it’s easier to slice a larger pie evenly than a smaller one. That’s the philosophy of what has to be the world’s most inspiring municipal leader on the subject of transportation equity, Enrique Peñalosa, the former mayor of Bogotá, Colombia.
Sometimes transportation equity is best studied in a place where inequality of all sorts is off the charts. That certainly describes Colombia’s capital city, whose seven million residents suffer from the greatest disparity between rich and poor in all of South America.
Differential access to transportation is, of course, not the only reason for the gap between Bogotá’s rich and poor, which was in place from the time of the city’s sixteenth-century founding as the capital of the New Kingdom of Granada on a narrow plateau bordered by the Andes Mountains on the east and the Bogotá River on the west. It certainly hadn’t improved much by the time Simón Bolívar liberated the Spanish crown colony in 1819. However, from the time that the city’s first transit system was opened, in 1884, transportation policy conspired to exaggerate Bogotán inequality. That system was a mule-drawn tramway that connected the city’s central square, the Plaza de Bolívar, with Chapinero, in the northern part of the city—not at all coincidentally where Bogotá’s most affluent residential and commercial districts were to be found, then and now.d
Unless it is carefully managed as a public resource, mass transit follows money. So when the old streetcar system was replaced in the 1940s by buses owned by private companies, the result was predictable. The best neighborhoods got the best service: relatively luxurious buses known as the ejecutivo, which prohibit standing riders. Everywhere else got the less expensive corriente. Even worse, the thousand-plus buses operated by the private companies did what competing companies do: they competed, which in Bogotán terms meant they frequently disrupted one another’s service. Since, by long tradition, the city’s sidewalks were regarded as extensions of the automotive arteries, they became parking lots for thousands of buses and, eventually, more than a million private cars.
Even so, from the 1950s through the 1980s, Bogotá continued to grow, not always for the best possible reasons. As the center of the Colombian drug trade, the city became notorious throughout the world not just for financial instability and inequality, but for crime. In 1993 alone, the city recorded more than four thousand homicides, which made it one of the most dangerous large cities in the world.
Then, in the middle of the 1990s, two successive mayoral administrations turned the city around, partly by improved policing and security, partly by better financial administration, but also by rethinking the city’s transportation system. The list of transportation innovations begun by Antanas Mockus when he was elected Bogotá’s mayor in 1995, and expanded by his successor Enrique Peñalosa from 1998 to 2001 (Mockus would, in turn, succeed Peñalosa, and serve until 2003), is nothing if not impressive.
The most significant, in terms of passenger miles, was the Bus Rapid Transit system known as the TransMilenio, a network built around a thousand 160-passenger articulated vehicles that covered the city’s longest and most traveled avenues on dedicated busways with elevated stations placed on road medians, with bus and station floors at a level for both convenience and safety. Tickets for the TransMilenio buy a full day’s travel, while the feeder routes for the BRT system are served by smaller buses on perpendicular roads to the main grid—and they are free, in order to make certain that the city’s less affluent are able to benefit from it. Small wonder that the system now carries nearly two million passengers per day, which is one of the highest usage rates in the world.
The TransMilenio, to be perfectly honest, is simultaneously a good news and a bad news story, both of which are instructive. The system’s biggest problem at the moment is overcrowding—bad enough that riders frequently can’t even exit the bus at their preferred stops—which is a reminder that the demand for public transit in Bogotá still exceeds supply. This is true even though the TransMilenio’s all-day fare—currently 1,700 Colombian pesos—is a very pricey option for low-income Bogotáns, who frequently earn less than 5,000 pesos daily. Even with the high fare structure, though, the system is severely underfunded. Though the TransMilenio was built as a private-public partnership, it still receives no subsidies from the municipal government.
In addition to shaky finances, the TransMilenio is a victim of poor execution. Like any Bus Rapid Transit system, it depends on decent roads, and the main artery used by the buses, the Avenida Caracas, was designed badly and maintained even worse. And, just to pile on, the system was also built without much in the way of state-of-the-art traffic engineering, which means that the schedules are the opposite of reliable, when they even exist (the system’s private bus systems have no published schedules at all).
Some of Bogotá’s other transportation innovations have been more successful. Mockus and especially Peñalosa were determined to make Bogotá a paradise for cyclists, and they very nearly succeeded. When Peñalosa took office, he immediately rejected a proposal to build what his engineers told him was the highest priority for the city’s infrastructure, an elevated highway costing some $600 million. Instead, out of a conviction that active transportation was not only more sustainable, but equally popular with every level of Bogotán society, he spent a fraction of that money on the ciclorrutas system: 234 miles of permanent bike paths separated from automobiles by curbs and the short posts known as bollards. Less useful for commuting, but even better loved by the city’s residents, he expanded the city’s ciclovías, which close seventy-five miles of Bogotá’s streets each Sunday and transform the temporarily car-free streets into plazas full of street entertainers, group exercise classes, and of course walkers and cyclists. It gets even better: ever since 2000, on the first Thursday in February, the entire city, rich and poor, goes car-free.
Peñalosa went further still. Those sidewalks that had been commandeered as de facto parking lots by the city’s traditional buses and cars? Peñalosa ordered them cleared, and then built a network of curbs and bollards to keep the vehicles off them permanently. He widened the sidewalks where he could—“skinnying” up the streets—and reduced the available street parking. To further discourage driving in the city center, and so attract more affluent riders into the city’s transit system, Bogotá restricted the number of private automobiles permitted during rush hour by 40 percent. You read that correctly. Depending on the last number of your license plate, four cars in ten were prohibited from Bogotá during peak travel times.e
The former mayor is now the president of the board of directors of the Institute for Transportation and Development Policy, and is rightly regarded as one of the world’s most articulate promoters of transportation equity. In an interview after a talk at Canning House in London (and again in his TED talk), Peñalosa observed, in words that I’d be proud to have on my own tombstone, “An advanced city is not one in which the poor can get around by car, but one in which even the rich use public transport.”
Penalosa’s goal is laudable but it’s a long way from assured. Although the revolutionary era that began when the first Millennials entered adulthood as car skeptics shows no signs of changing direction, and the pace of innovation in sustainable, active transportation is, if anything, accelerating, the road ahead is nonetheless still under construction, and some obstructions are predictable.
The first is political. Like everything else in this feverishly partisan era, transportation policy is a blood sport for both self-described progressives and conservatives. As Houston has proved, once a fairly straightforward decision about infrastructure gets turned into a proxy debate about competing myths about America, rationality flies right out the window. I am guilty of this myself. I am inclined to assume, for example, that knowing how frequently someone uses public transit or rides a bike tells me what that someone thinks about gay marriage, climate change, and gun control.
It’s not that this kind of thinking is completely wide of the mark. Bill Bishop in The Big Sort describes how Americans have been grouping themselves into like-minded communities for generations now, and how one of the key markers for almost every political choice is where they choose to live. Or, more accurately, the density of the communities they choose. During the 2012 presidential election, 98 percent of America’s most densely populated counties voted for Barack Obama and 98 percent of the country’s least densely populated counties voted for Mitt Romney. It works at almost every level of granularity: until a community—a county, a city, a town, or even a voting precinct—reaches a density of about eight hundred people a square mile, it’s as reliably Republican as Fox News. Once it exceeds that number, though, the voting patterns do a somersault. Anywhere under eight hundred people a square mile, there’s a two-thirds chance that a randomly selected voter went Republican; above it, that hypothetical voter pulled a Democratic lever two-thirds of the time. As the political prediction machine Nate Silver of 538.com tweeted in 2012, “If a place has sidewalks, it votes Democratic.”
It’s not totally obvious whether people vote a certain way because of where they live, or whether they move to places where everyone votes the way they do. What is obvious though is that all the elements of a Street Smart transportation system depend on density. At first glance, this would appear to be a giant advantage for a Street Smart future, since every demographic indicator shows that America and the world are headed for a much more urbanized future. Between 1970 and 2000, the world’s urban areas grew by about 22,300 square miles, but in the three decades between 2000 and 2030, they are expected to grow by 590,000 square miles, and house nearly one-and-a-half billion more people than today. All those Millennials and Boomers migrating to big cities are just the leading edge of an avalanche.
On the other hand, all that action is causing a powerful reaction. When Enrique Peñalosa lost his bid for reelection in 2000, he was followed by three successively more conservative administrations, and it’s not too much to describe what they’ve done to some of his signature transportation initiatives as sabotage. In the United States, the reactionaries pushing back the hardest on urban public transportation systems are led by the billionaire brothers David and Charles Koch, and their umbrella advocacy organization, Americans for Prosperity. Perhaps that’s not surprising. In September 2014 Tim Dickinson in Rolling Stone described what the brothers’ businesses were: “Koch-owned businesses trade, transport, refine and process fossil fuels.”
This book is way too short to document all the silliness and conspiracy-mongering funded by AFP. But no matter how much time they spend on climate-change denial, repealing the Affordable Care Act, or attacking Agenda 21, the nonbinding United Nations blueprint for sustainable development, AFP and the Koch brothers–funded Reason Foundation always seem to find a few idle hours each day to oppose public-transit investment. In 2014 alone, they spent hundreds of thousands if not millions of dollars undermining a program of dedicated transit lines in Nashville; forbidding the city of Indianapolis from even studying a light rail system; fighting—and, happily, losing—battles opposing the Washington Metro’s expansion into Loudon County, Virginia, and Los Angeles’s Exposition Line rail system; and killing Florida’s plans for a high-speed rail system, which had been overwhelmingly approved by the state’s voters. Urban populations may be growing, but they’re going to have to fight for improved public transit so long as the Koch brothers can continue writing checks.
On the other hand, the partisan divide between transit-loving liberals and car-adoring conservatives may not be as wide or as deep as it first appears. It’s true that polls consistently show substantial differences in transportation and housing preferences between liberals and conservatives. In a 2012 Pew Research Center study, three-quarters of self-described consistently conservative voters said they would opt to live where “the houses are larger and farther apart, but schools, stores and restaurants are several miles away,” while only 21 percent of consistent liberals said the same. Meanwhile, 71 percent of liberals (but only 22 percent of conservatives) would choose communities where “the houses are smaller and closer to each other, but schools, stores and restaurants are within walking distance.” Conservatives like big lawns; liberals like walkability.
And they vote accordingly. Just about every big city in the United States consistently votes Democratic. Even in Republican states like Missouri, more than 80 percent of the electorate in a city like St. Louis votes Democratic.
However, the exceptions to this rule are the really interesting ones. Two of the biggest cities that voted for Mitt Romney in the 2012 presidential election were Salt Lake City and Oklahoma City. Despite that, as we’ve seen, both cities recognize the critical importance of building the elements of a Street Smart transportation system, from walkable downtowns to multimodal grids. Ideologically driven politicians and think tanks can fulminate all they want about the creeping dangers of European-style urbanism in the land of the free and the home of the brave. But mayors and city managers all over the country, whatever their political affiliations, can’t afford to see transportation policies in those terms. They know that the only future that will keep their cities vital and attractive to Millennials and the generations that will follow them isn’t reachable without streetcars, sidewalks, and bike paths. As a result, I’m not terribly worried about the increasingly desperate tactics of the reactionaries to disrupt the ongoing revolution in America’s transportation networks.
Nor am I concerned too much about the price of oil, which spent the second half of 2014 behaving like a skydiver with a bowling ball strapped to his back instead of a parachute. In six months, the price of a barrel of oil fell from $115 to less than $60, and was forecast to stay somewhere under $70 for at least another year. This is a very big deal in a lot of ways. It puts a huge amount of pressure on big oil-exporting nations, from Saudi Arabia to Russia to Venezuela. It will probably cause a reassessment of hydraulic fracking and nuclear-power-plant construction, and is a gigantic bonus for almost the entire US economy.
One thing it is not, though, is something that is likely to change the direction, or even the slope, of all those graphs showing a decline in annual vehicle miles traveled. When the really sharp decline in VMT began in 2004, the price of a barrel of oil was less than $40 a barrel. Five years later, when the VMT decline had become obvious to every transportation planner in the country, it was about $44.f That price didn’t last, but then there’s little reason to believe this one will, either. For a little while, at least, car owners are going to be paying less for every mile they drive, but it’s hard to see why they’ll be putting too many more of them on their odometers. In addition, the average driver pays about $9,000 a year to own a car. Saving a few hundred bucks a year on gas won’t change the financial decision tree much.
That doesn’t mean there’s no risk of any bad outcomes. Given the difficulties most of us have with distinguishing between short-term appetites and long-term good sense, there’s a chance that lower oil prices will lead to some poor decisions on infrastructure investment.
Even if the price of oil doesn’t distort investment decisions, it’s not as if we were making the most efficient decisions on infrastructure before. Though there is actually a lot to recommend in the American Society of Civil Engineers’ analysis of the sorry state of all aspects of the country’s infrastructure—we really do need to upgrade our waste-water treatment plants and our internal waterways—the Report Card makes some serious errors in urging greater investment in big parts of the transportation system. Spending hundreds of billions of dollars on expanding and restoring roads when Americans are driving fewer miles each year is a very expensive exercise in nostalgia, one that could easily turn our fifty-year-long mistake into one lasting decades longer.
This is especially true when the words deficient or obsolete are used to describe elements of the transportation infrastructure. Most people believe that when a professional organization of civil engineers says that more than sixty-five thousand American bridges—11 percent of the total—are “structurally deficient,” they mean that they are in imminent risk of collapse. Few actually are, though. Most of the structural deficiencies are part of normal wear and tear, and simply indicate a need for rehabilitation of one or more components. (There is a class of structurally deficient bridges that does worry me: the ones that are fracture critical. These are spans—some eighteen thousand of them, all built more than forty years ago—designed without what engineers call complete redundancy; this means that if just one critical beam or a single connecting joint fails, the entire bridge can collapse.)
The bigger problem, though, is the more than eighty thousand bridges that the ASCE calls functionally obsolete. This is another engineering term that just means a particular bridge isn’t wide enough, or robust enough, to carry the maximum amount of traffic, including the biggest trucks, under conditions approaching free flow. This was the argument made by the federal government during the tug-of-war over the Williamsburg Bridge. Had we decided to replace the bridge, New York would have had to spend three-quarters of a billion dollars on a bridge whose primary effects would have been to destroy existing neighborhoods on both sides of the East River and put even more cars and trucks on Manhattan’s streets.
This doesn’t mean that we shouldn’t be urgently repairing needed bridges, or roads, or railroads. The key is recognizing which ones are needed, and which not. And how much we have to spend to do the right repair.
The last bit is critical. More than forty years of participating in the political process by which we allocate transportation investment has taught me that it’s far easier to secure money for new infrastructure than for maintenance of the stuff we already have. Some of the reasons are found in human nature: everybody loves a ribbon-cutting, and mayors and governors are a lot happier seeing their names on the congratulatory signs that accompany a new bridge than on ones about repainting the understructure of an old one, even though the bang-for-the-buck equation almost always favors the latter. Other reasons are statutory. When I was responsible for New York City’s 840 bridges, the annual budget that came with the job was around $400 million, almost all of it for capital expenditures. That is, I had the equivalent of the entire municipal budget of a city the size of Spokane for building bridges, but next to nothing for painting or repairing them.g It’s not exactly unknown for manipulative city budget analysts to force an agency to forgo maintenance of infrastructure for exactly this reason.
If there is a lesson from all the preceding chapters in this book, it’s that a transition to a Street Smart transportation infrastructure isn’t just aligned with the changing preferences of young, and not-so-young, Americans. It’s that the transition doesn’t need to cost any more than our current resources allow, and probably a whole lot less. Unbuilding, as with the West Side Highway or the Embarcadero—or, even better, never building—is far cheaper than building and then maintaining forever. There is reason for concern that we haven’t yet learned that lesson completely, but also reason for hope that the next generation of transportation professionals understands that their primary objective will be improving access and mobility, not increasing capacity. If they do, the costs of new infrastructure will be a lot lower than anyone expects, and far lower than staying on the path we’ve been traveling for the last fifty years.
More concerning to me is a very specific kind of technology, that of autonomous driving. In 2014, a team of researchers from the Rudin Center for Transportation Policy and Management at New York University tried to imagine the impact of technology on the mobility in four US metropolitan areas—Boston, Atlanta, Los Angeles, and northern New Jersey—in the year 2030. They used an approach for describing alternate futures originally developed at the University of Hawaii in the 1970s, one that assumes that all possible future narratives take one of four basic shapes:
• Present trends continue on the path they’re at, either growing exponentially or linearly.
• Things collapse. Critical systems fail and others deteriorate because of predicted and unpredicted events.
• The future runs into a wall. Limits on key resources place a constraint on growth, and a sustainability model of slower or zero growth takes its place.
• An unpredictable transformation occurs. Some disruptive technology appears, changing the direction of, well, everything.
In transportation, the ultimate disruptive technology may well be upon us: driverless cars. Though it may not be correct to call it unpredictable. As far back as the 1920s, a radio-controlled car from the now-defunct Achen Motor Company navigated its way through downtown Milwaukee. In 1940, Norman Bel Geddes, the same architect and industrial designer hired by Walter O’Malley to imagine an Ebbets Field of the future, anticipated that the cars of 1960 would
have in them devices which will correct the faults of human beings as drivers. They will prevent the driver from . . . turning out into traffic except when he should. They will aid him in passing through intersections without slowing down or causing anyone else to do so and without endangering himself or others.
For the next five decades, companies like RCA, General Motors, Mercedes-Benz, and others worked to bring Bel Geddes’s vision to life. For most of that time, autonomous vehicles were conceived as part of a system that traveled on dedicated roads or tracks, rather than streets, and went by the name of Personal Rapid Transit, or PRT.
PRT is generally used to describe a network of small, driverless electrical vehicles—pod cars—traveling on elevated guidewaysh containing sensors and switches that can, in combination, offer point-to-point travel nearly as flexibly as an automobile, but as safely and efficiently as a subway or streetcar. PRT has had a number of champions over the last fifty years (although more detractors), and a few fairly successful small-scale systems have even been built. One of them, at London’s Heathrow Airport, uses eighteen vehicles traveling on two-and-a-half miles of track to transport travelers among three terminals. The Morgantown campus of the University of West Virginia has a similar system, and others are planned from Korea to Sweden. Still more remain under consideration in Mountain View, California, and Boston.
However, the scalability problems of PRT seem almost insuperable. Princeton’s Alain Kornhauser, one of the technology’s earliest and most articulate proponents, calculated the requirements for a PRT system that could actually replace the twenty-five million daily trips taken each business day by private automobile in the state of New Jersey. The numbers are staggering: 215,000 electric vehicles traveling among more than eight thousand interconnected stations, at an initial capital cost that would exceed $200 billion.
By far the largest part of that enormous price tag comes from constructing the elevated guideways on which the system depends: for New Jersey alone, Kornhauser’s system would require more than ten thousand miles of them, a quarter the length of the entire Interstate Highway System, which makes me think the $200 billion price tag is still way too low. For a long time, such guideways seemed absolutely necessary for segregating driverless vehicles from existing streets and so avoiding the danger of collision. If a way could be found, though, to run driverless vehicles just as safely on existing streets, the system’s cost starts to become affordable, not least because, unlike a traditional PRT network, driverless cars can share the road with traditional automobiles. That would mean that a system using them wouldn’t need to be fully built out before it becomes useful.
That was the idea behind the Autonomous Land Vehicle project, which was initiated in the 1980s by the US Department of Defense Advanced Research Projects Agency, who enlisted a hundred different university engineering departments in a contest to produce a robotic vehicle. Some came very close. Navlab 5, one of a series of vehicles developed at Carnegie Mellon University, drove from Pittsburgh to San Diego, a journey of 3,100 miles, more than 2,900 of them driverless.i By 2005, the ALV program gave way to a new DARPA initiative, the Grand Challenge, won by “Stanley,” a car created by Stanford University and Volkswagen’s Electronics Research Laboratory. Like the sophisticated GPS signaling and open-access transportation data that make both multimodal transit systems and car ride-matching services like Uber and Lyft viable, the newest generation of driverless cars combine incredibly precise real-time mapping algorithms with remote sensing systems that use radar, sonar, and lasers to “see” other vehicles as well as obstacles.
By 2010, every major automobile manufacturer was heavily invested in autonomous driving technology that could be used on existing roads. Part of the enthusiasm for the technology is that components of the technology required for a truly driverless car are valuable on their own. Adaptive cruise control, for example, a system that can detect distance between one car and another and modify speed accordingly, is already available on a number of luxury automobiles. So is autonomous steering, which can do the same for lane keeping, as are systems that can drive a car into a multistory parking structure and ease itself into a space directed by a smartwatch rather than a driver, and even solve the bugaboo of drivers everywhere, parallel parking a car in a space only ten inches or so longer than the car itself. In the view of former PRT advocate Alain Kornhauser, who is now convinced of the practicality of street-useful driverless cars, the beauty of these technological improvements is that, because they increase driving safety, they even have the potential to be self-financing: so long as collision avoidance and other autonomous driving modules cost less than the potential liability from future accidents, it will be in the interest of automobile insurance companies to pay for them. Even better: so long as more autonomy equals more safety, there is no point where the cost of the technology exceeds its added value.
The most prominent player in the world of autonomous driving, though, isn’t Allstate or Geico. It isn’t Mercedes-Benz or Ford, or even Tesla. It’s Google.
The Google Self-Driving Car is a project that the Internet giant saw as a natural outgrowth of its existing mapping software, particularly the technology from Google Street View, which stitches together panoramic photos of more than five million miles of roads in more than forty countries. Google’s versions of the driverless car—refitted Toyotas, Audis, and Lexuses—combine real-time access to all that data with a laser rangefinder that creates and refreshes three-dimensional maps of the area immediately around the car. It has so far succeeded in a dozen different road tests, comprising more than seven hundred thousand autonomous miles without a single self-caused problem (one car did get rear-ended; not, one hopes, by another autonomous vehicle). Though the company admits to a number of limitations to the existing technology, including bad weather, the Google car has done a spectacular job promoting the potential of autonomous driving. For people who believe in the never-ending bounty of digital improvement it seems only a matter of a few years before Google solves the remaining technical obstacles in the path of truly autonomous driving.j (At that point, Google, which invested more than $250 million in Uber back in 2013, will be able to launch its new subsidiary, which I call Goober.)
Lots of people are sold. I’m not sure who first pointed out that predictions are difficult, especially about the future,k but whoever it was had a point. Here are a few popular notions about the likely consequences of autonomous driving technology:
In 2012, the Institute of Electrical and Electronic Engineers predicted that driverless cars will account for “up to” 75 percent of the vehicles on the road by 2040. (You have to love the way two little words like “up to” can cover the asses of so many engineers.) Alberto Broggi, a member of IEEE and professor of engineering at the University of Parma, believes they will be capable of speeds “up to one hundred miles an hour by 2040” (emphasis added).
Meanwhile, the marketing consultancy Navigant Research predicts that by 2035, annual sales of autonomous vehicles will be more than 95 million. At the other end of the spectrum, Columbia University’s Earth Institute calculated, in 2013, that, by combining the Internet, advanced propulsion systems, vehicle sharing, and driverless cars in Manhattan alone, nine thousand driverless cars could replace thirteen thousand cabs, with both shorter wait times and faster travel, all at a per-mile cost of one-tenth of that incurred by present-day cabs.
The consulting and accounting firm PwC goes even further, predicting that the number of traffic crashes occurring in the United States every year could drop from 10.8 million to 1.1 million (just switching from driving to transit also achieves a better than 90 percent reduction in crash probability on a per-person per-mile basis), that congestion-based “wasted fuel” could fall from 1.9 billion gallons annually to 190 million gallons, and that the privately owned US vehicle fleet could collapse from its current level of 245 million to only 2.4 million. Not to be outdone, the prognosticators at Morgan Stanley predict that the United States would save $158 billion annually in fuel, $563 billion from reducing automobile crashes, and $422 billion in productivity improvements (from allowing people to do useful work while traveling in their driverless cars), for a net addition to the American economy of more than $1 trillion a year. With all that, it’s not too surprising that many advocates for driverless cars argue that any investment in old-fashioned, capital-intensive transportation infrastructure is foolish, given the very real possibility that such investments would be obsolete before they were even operational.
I hate to be a killjoy. But there are a few speed bumps (sorry) in the path of those predictions.
First, while Google Maps and Google Street View are both incredible pieces of software, the level of data collection—and especially the updating requirements—required of maps needed by the computer operating a Google car is at least an order of magnitude more complex. All those successful test drives have been performed using a dedicated vehicle that preceded the autonomous car on the route in question, uploading every physical object on that route into the car’s memory immediately before the journey. This isn’t quite as big a con job as a mentalist who has confederates checking out marks before a performance, but neither is it a system that can be scaled up for real-world use. Remember: we’re talking not just roads, but driveways, parking lots, ferry terminals, and basically anywhere a car can already travel.
Then there’s the uncomfortable reality that the map is not the territory. No matter how frequently it is updated, no mapping software can keep up with every new lane marking, or pothole, or construction site, or utility truck blocking a road. As I write this, most street markings in the Northeast are covered by snow and ice, and snow banks often require driving illegally into lanes reserved for oncoming traffic. (By the way, who gets the ticket if a driverless car breaks a rule and how does it know to pull over for a cop?) If an autonomous vehicle doesn’t know a traffic light is in the road, it can’t obey it—and temporary portable traffic lights are moved to new intersections every day. And that’s without even mentioning the so-far-untested ability to distinguish between a piece of newspaper in the road and a rock; a police car’s flashers and a turn signal; or—and this is the scary one—a deer, a dog, and a five-year-old child.
Nonetheless I am convinced the driverless car will be safer overall than those driven by even more fallible humans. However, it might not be the most efficient way to keep people out of the hospitals, or the morgues.
More than eighty years ago, a cartoonist named Rube Goldberg became famous for his drawings of needlessly complicated gadgets that performed the simplest tasks in the most convoluted ways imaginable, using levers, cranks, balls rolling downhill, and balloons rising upward to pour a glass of milk, for example. In a way, driverless cars seem like a Rube Goldberg approach to getting from here to there. For short trips, walking and biking are safer and healthier than Google’s most ambitious vehicle; for many longer ones, so is transit. Not only does switching from driving to transit also achieve a better than 90 percent reduction in crash probability on a per-person per-mile basis, it also costs energy—in a good way. Transit riders use more than 20 percent more calories than drivers on a per-trip basis, which gives buses, subways, and streetcars a giant health advantage over cars. In fact, after five years of taking transit, the obese percentage of a given population—those with a Body Mass Index greater than 30—drops by more than half.
And, as long as cities create plazas and piazzas where cars are banned but not people, self-driving cars offer no advantage, even without recognizing the mathematical impossibility of moving thousands of people through a city center in single-occupant vehicles. This doesn’t mean there isn’t a place for cars, with or without laser-rangefinders and GPS mapping. In less dense parts of cities, suburbs, and rural areas, all the safety aspects developed by automated cars make sense.
In some ways, the driverless car is a natural next step following all the technological and demographic changes that contributed to the original Millennial-led driving revolution that is the subject of this book, especially the information oversupply that made smartphones into a tool for transportation planning. Driverless cars are also, in their way, a new army on the battlefield over which the original war for right-of-way was fought nearly a century ago, in which roads were transformed from multipurpose commercial and social real estate into single-use arteries for automobile travel. Driverless cars are unarguably transportation anywhere, and anytime. And, to the degree that driverless cars will be available to all levels of society, they’re transportation for everybody, too. They can even coexist, peacefully and profitably, with streetcars, commuter rail, bike paths, and subways.
Still, it isn’t clear whether that new army of driverless cars is a Street Smart ally or opponent. There are any number of potential benefits to be found in a world dominated by driverless cars, including the possibility of turning nine out of every ten vehicles currently on the road into planters. As hardly needs underlining, the environmental and social benefits of taking 90 percent of existing cars off the streets are almost incalculably high. Moreover, given that some 93 percent of the six million automobile crashes in the United States every year are at least partly due to human error, that’s something like thirty thousand deaths that might be prevented by getting human beings out of the business of driving. But because driverless cars are possible only insofar as they use existing streets and roads, they are still subject to the same hard cap of the vehicles in motion calculus described in Chapter 7. Because a computer’s reflexes are better than yours or mine, autonomous vehicles can pack roads more efficiently than traditional ones; the cap is higher, but it doesn’t disappear. The rosiest scenarios for autonomous driving, the ones that forecast a precipitous drop in the number of cars on the road, assume that most of those cars won’t be personal vehicles but part of fleets of what we might call A-taxis. This means that virtually all of them will be in motion, virtually all of the time. In cities, particularly the densest parts of cities, driverless cars may be a recipe for constant gridlock.
However, even if driverless cars can, theoretically, reduce traffic delays because more of them can travel safely on a given stretch of road, this doesn’t mean that they will. One reason for the anticipated appeal of driverless cars is that they are expected to mimic travel by train: smooth enough for reading, or working, or Internet gaming. But in order to do this, they would have to also mimic the (slower) acceleration profile of trains—and when they do, according to a recent simulation by a group of researchers at University College London, they don’t improve travel times. They increase them. A group of simulated driverless cars negotiating a typical urban intersection at the same (slow) acceleration of a commuter train increases the time needed to cross the intersection by anywhere from 36 percent to more than 2,000 percent. If you want to browse the Internet while commuting, and still want to get to work on time, trains look like a much better option.
There are other reasons to be suspicious of the brave new world represented by Google’s self-driving cars and others of similar ambition. On a purely personal level, I’m a little taken aback by the promise that autonomous vehicles will be able to collect you at your front door and deposit you at the front door of a supermarket or shopping mall—or even at your desk or workstation—without your feet ever touching the ground. In the Disney movie Wall-E, spaceship-bound refugees from an Earth destroyed by environmental catastrophe are so well cared for by their robot transportation devices that hardly anyone even stands up anymore, with the result that the universe’s entire remaining population of Homo sapiens is morbidly obese. This, it seems to me, is not a particularly utopian future—one in which fewer people die from crashes, but more get hypertension and diabetes at ever-younger ages. Also, virtually all the really dramatic predictions about the benefits of driverless cars assume an entirely driverless network—one in which no one drives, and for which virtually driving is done autonomously. This is a nontrivial point: a system that is “only” driverless on expressways, for example, isn’t going to change behavior in large ways, since most trips are less than ten miles in length. And don’t get me started on trying to figure out who gets sued in the event of a collision between autonomous cars.
Maybe more plausibly, others have wondered whether autonomous cars, by reducing the pain and misery associated with driving, will therefore make it more appealing—so appealing, in fact, as to reverse the centripetal phenomenon that is now drawing more and more people back into densely populated cities from the sprawling suburbs that attracted their parents and grandparents after the Second World War. In that scenario, a new generation of commuters will be so happy to enter a driverless vehicle—one that allows them to watch movies, read books, or catch up on e-mail without ever having to worry about other drivers, traffic jams, or even missing that exit on Route 124—that they will be quite content to accept commutes that run into hours each day. Why not? It’s not torture anymore but a chance to binge watch all those episodes of The Sopranos on your high-definition tablet. Sprawl would be a natural and inevitable consequence.
It depends, I think, on whether we believe that the revolution in driving behavior ignited by the Millennials is largely a reaction to the costs of automobile commuting in terms of money, time, and discomfort. If that’s the case, then anything that removes or lessens those costs does indeed have the potential not just to adjust the curve but to reverse it.
But I’m unpersuaded. I believe that the behavioral changes we’re seeing are less a matter of the pain and misery of commuting by car than they are about the pleasures of living in a walkable, accessible community. If that’s the case, then autonomous cars would be exactly the kind of beneficial technology that could actually help us to return to the kind of human-scale, livable transportation system that existed before we decided to subsidize sprawl and penalize density.
Only better. While I’m personally nostalgic for the kind of streets on which I grew up, I also know it’s important not to sentimentalize them too much. The streets of late-nineteenth-century Brooklyn or early-twentieth-century Los Angeles were walkable, but they were also dirty, overcrowded, and unsafe for both vehicles and pedestrians. It’s worth remembering that the Brooklyn Dodgers were originally nicknamed the “Trolley Dodgers” in 1895, by which time electric trolleys were killing between thirty and fifty Brooklynites annually. One of the beauties of the Street Smart program is that it uses groundbreaking new technologies to upgrade the transportation systems—roads and rails, sidewalks and bikeways—in ancient, or at least centuries-old, cities.
But there’s another advantage to smart streets. Smart street principles—narrower, traffic-calmed thoroughfares; enough density to promote walkability; Internet and GPS-enabled wayfinding; and multiple choices for both transportation modes and connections—can be economically and successfully applied anywhere.
This is more important for transportation equity, even for democracy, than is generally understood. The neighborhoods and cities that are leading this particular revolution have been, in general, wildly successful, but one consequence of that success is that the cost of living in them has increased dramatically. The demand for walkable, street smart neighborhoods with good public transit is, at present, far exceeding the supply (and it wasn’t as if homes in San Francisco or Manhattan weren’t already in great demand). To the degree that we smarten up neighborhoods in places where Millennials, in particular, already want to live, we run the risk of making them unaffordable to all but the most prosperous of them. If the revolution were simply to promote more active transportation and easier-to-use transit in the wealthiest American and European neighborhoods, it will have fallen short of its promise.
That’s why the preceding chapters made a point of visiting places like Charleston and Columbus and Oklahoma City. Density is just as prized in the upper Midwest as it is on the East Coast. Houston can benefit from a rebuilt and walkable downtown just as much as Pasadena. Most of all, the digital systems that make bus routes, rail lines, and even ride-matching car services transparent to their users work just as well in the cities of the Wasatch Range as in Northern California. The Street Smart revolution truly is everywhere, all the time, and definitely for everybody.
a These are popularly known in the transportation racket as “TEA-21” and “SAFETEA.” They were succeeded, in their turn, by the “Moving Ahead for Progress in the 21st Century” bill in 2012, abbreviated as MAP-21. I have occasionally wondered if congressional staff members are tested for their talents at acronym forming at their initial job interviews—and whether the people who get hired are the ones who score the lowest.
b Actually, Plessy had only one African American great-grandparent, but according to the enlightened laws of Jim Crow Louisiana, that made him black.
c From the Department of Unexpected Consequences: not only did the expected traffic congestion never appear, but so many Brooklynites used the free ferries as an easy way to travel to Manhattan that IKEA now charges $5 to any rider without a receipt. Others use the esplanade that IKEA built for its water taxis as a park.
d The original tramway was electrified in 1910 and ran until 1948, when Bogotá, like so many other cities in North and South America, eliminated its streetcar system.
e Some argue that this has had an unforeseen consequence: wealthy Bogotáns purchasing additional cars (with different last numbers on their license plates) to evade the restriction.
f In both cases, even the inflation-adjusted numbers are about 20 percent lower than the price at the end of 2014.
g One of my smaller triumphs as a public servant was getting things like paint classified as a capital expense.
h One system currently in development in Masdar City, just outside Abu Dhabi, will run underground.
i More or less. The car steered itself, but humans controlled throttle and brake, out of a perfectly reasonable concern for safety.
j As of this writing, another technological behemoth, Apple, is rumored to be developing an automobile that may be self-driving.
k It’s been attributed to everyone from the Nobel Prize–winning physicist Niels Bohr to Yogi Berra, and is almost certainly the only time those two giants of the twentieth century have been confused with one another.