How Traffic Jams Are Made In City Hall

The bad logic and failed policies of transportation planners

If you want to know why so few people use mass transit, meet Sue, a college administrator in Minneapolis. If anyone would use transit, Sue would. She's single, she lives in a condominium, and she can afford any additional out-of-pocket expense. She could use her city's Hiawatha Line, a light rail route newly completed at a cost of $715 million. But she doesn't, although she feels guilty about it. That's because her car gets her where she needs to go. Faster.

According to the U.S. Census Bureau, the typical driver in America's metropolitan areas takes 21 minutes to get from home to work. If you take public transit, the average commute stretches to 36 minutes. That's 71 percent longer. Workers in the New York metropolitan area have the longest commute: There it takes an average of 52 minutes to get to work, even though the New York-New Jersey-Connecticut mass transit systems are among the most extensive in the nation.

Minneapolis-St. Paul is about average. The typical commuter takes 21 minutes to get to work by car or 32 minutes by public transit. Congestion can be pretty bad: The average driver in the Twin Cities spends 43 hours-more than an entire work week-stuck in traffic every year. According to the Texas Transportation Institute at Texas A&M University, that costs Twin Cities drivers almost $1 billion in wasted time and fuel. But mass transit takes even longer, and it isn't as flexible as a car when it comes to picking where and when you'd like to go. Is it any wonder Sue drives to work rather than taking the bus or train?

The U.S. Department of Transportation puts the yearly cost of congestion at $168 billion. But the planning gurus who are supposed to solve our transportation problems are in the grip of transitphilia and autophobia; their beliefs about how cities and transportation work are grounded more in nostalgia than in a realistic view of the world we live in now. The public policies they design and try to enforce make it harder for us to get to work, pick up our kids from school, or go shopping. They are deliberately fostering congestion. In the words of David Solow, head of the Metrolink commuter rail in Southern California, congestion is "actually good" because "it drives people out of their cars."

Keeping Minneapolis Congested

Every major urban area in the country has an official bureaucracy responsible for planning roads, highways, and mass transit. It has to; it's required by federal law.

Minneapolis has one of the more competent planning agencies. The Metropolitan Council-or Met Council, as locals call it-has at least acknowledged the importance of congestion and has tried a few innovative ways to address it. Unfortunately, its solutions will have minimal impact on the problem. It provides an instructive example of how poorly even our better regional planning agencies are addressing one of the most important policy problems they face.

The Met Council has some extraordinary powers. Established by the Minnesota legislature in 1967, it has legal responsibility for managing the Twin Cities' sewers, parks, transportation, aviation, and land use planning. But the primary focus of its huge organizational bureaucracy is transportation. Of its 3,718 employees, 73 percent do transportation-related work, spending three-quarters of the agency's annual budget.

The council says it aims to enhance "transportation choices" and to "improve the ability of Minnesotans to travel safely and efficiently through the region." So far, so good. The council goes even further: "To a growing number of metro area residents, traffic congestion ranks as the No. 1 livability issue. It affects the length of their daily commute, the times of day they choose to make trips, and the amount of time they sit in traffic, even where they choose to live and work."

But is the Met Council really focused on reducing congestion? Its stated goal isn't to solve the problem; it merely calls for "slowing the growth in traffic congestion and improving mobility" (emphasis added). In other words, traffic will continue to get worse, just not as much worse as it would if the council did nothing. The Met Council also has priorities besides congestion: reducing the number of people living in single-family homes, preserving open space, limiting sprawl-and increasing transit use.

During the next 10 years, the Met Council is planning to invest $4.2 billion in the highway system and $1.4 billion in transit facilities. In other words, the region's primary transportation planning agency has decided to spend 25 percent of its budget on mass transit. But transit accounts for just 2.5 percent of all trips in the region, whether they're for pleasure, taking kids to school, going to the supermarket, or commuting to the office. Less than 5 percent of the Minneapolis-St. Paul region's population uses public transit to get to work, and that share is declining: According to U.S. Census statistics, the number of passengers using mass transit increased slightly in absolute terms between 1990 and 2000, but its market share fell by 12 percent.

The Met Council hopes to double bus capacity by 2030 and greatly expand its light rail line and commuter train system. It also intends to boost transit use from 74.9 million passenger trips per year to 150 million by 2030, even though the current trend projects virtually no growth in use and even though transit lost market share from 1990 to 2000, according to the Census Bureau's decennial data. The Met Council expects 574,625 new jobs to be created in the area by 2030. But even though the vast majority of Minneapolis-St. Paul's population travels to work by car, the planners improbably expect per capita road use to decline.

The council does plan to expand the road system. It will add 300 additional lane-miles of freeway, or about 12 lane-miles per year. That works out to about three miles of a two-lane (in each direction) freeway each year. That's well below the expected growth in travel demand.

The net result? Without road improvements, highway congestion is expected to increase from 28 hours annually per traveler in 2001 to 40 hours in 2030. With the improvements, congestion should "moderate" to 37 hours in 2030. Congestion would be 32 percent higher than in 2001, rather than 42 percent higher without the improvements. "Just to keep pace with these [highway] needs," the council's 2030 Regional Development Framework says, "would add $4.7 billion to current plans for the next decade" above the currently planned spending.

For most regional planning agencies, automobility and congestion relief simply are not high on the priority list. Sometimes they aren't on the list at all. Portland, Oregon, distinguished itself among its peers when it made a conscious decision in the mid-1990s to let congestion approach gridlock because it feared that otherwise fewer people would use the transit system. The drive to reduce sprawl creates a conflict of interest, too, since congestion relief makes it easier to commute long distances.

To make "more effective use" of the road system, the Met Council believes it has to get people out of their cars. That's unfortunate, especially since the agency admits congestion is many residents' "No. 1 livability issue." The council is spending 25 percent of its transportation funds on a solution that, at most, might improve the quality of life for 5 percent of the population, and it will do nothing for people like Sue. Even transit users might not be better off, since they will be spending more time commuting than if they used a car. Drivers will definitely be worse off. They will be spending much more time stuck in traffic in 2030 than they did in 2006.

If Minneapolis has one of the best planning agencies, what are the others like?

Fiddling While Atlanta Burns Gas

According to the Texas Transportation Institute, Atlanta is the nation's fourth most traffic-clogged metropolitan area, measured by the amount of time stuck in traffic. Its residents crawl and wind through more than 5,000 congested lane-miles each day. Most of these congested roads are arterials and collectors-local roads that let residents navigate short trips around town or their neighborhood or that take them to major highway interchanges. An analysis by the Atlanta Regional Commission of 75 intersections found that 60 were "deficient"-that is, they performed below engineered standards-during the morning rush hour and 68 were deficient during the afternoon rush. The freeways are even worse off: Almost 60 percent of Atlanta's interstates are congested, twice the incidence for local roads.

The South isn't normally seen as a hotbed of progressive government, but the Atlanta Regional Commission was the nation's first government-supported multi-county planning commission. The Atlanta Chamber of Commerce hatched the idea in 1938, and it became official with an act of the Georgia General Assembly in 1947.

As clogged roads slowly choke Atlanta's economy and its quality of life, traffic reduction should be the commission's No. 1 priority. And the commission appears to take its role seriously. During the next 25 years, it plans to spend $57 billion on transportation projects even if the federal, state, or local governments don't cough up more money. (Presumably the funds would come from tolls or other user fees.) Its plan, however, assumes that vehicle miles traveled per person-a common measure of travel demand-will fall by 5 percent and that average travel time won't change. The plan anticipates "significant improvement in congestion and travel times" along the corridors targeted for investment, saving billions of dollars through improved efficiency and productivity.

The commission has some reason for optimism. Travel demand appears to have fallen in Atlanta from a peak of 35 vehicle miles traveled per person each day in 1998 to 31 vehicle miles in 2002. Total demand has increased-from 109 million vehicle miles traveled in 1998 to 113 million in 2002-but that's because population has grown so much. (The number of people living in the Atlanta area increased by more than 200,000 during the same four years.) Each person is driving slightly less, but since there are so many more people, the roads are getting more use than ever.

Naturally, congestion increased during this period too, reflecting the increase in travel demand without a similar investment in roadway capacity. Travel times to work also increased, according to the U.S. Census Bureau, rising 24 percent during the 1990s to 30 minutes in 2000. The commission reports that this was the largest increase in the nation.

So congestion is increasing, even though demand seems to be moderating. And local policy makers aren't expecting much more help from the federal, state, or local governments.

What's Atlanta's plan? Roadway expansion will get $8 billion. Car pool lanes will get another $5 billion, bringing the total pavement capacity building budget to $13 billion. The commission has slotted another $14 billion for nontransit operations and maintenance. These efforts will add 2,000 additional miles of arterial and collector roads and 300 miles of new freeway lanes. Another $3 billion is slated for improving the management of the road system.

Atlanta also believes that improving traffic signal timing to smooth out traffic flows, using meters on entrance ramps to prevent too many cars from entering the freeway at the same time, and similar measures that "manage" travel demand will reduce delays on local roads by 25 percent and increase freeway speeds by a similar magnitude. If those plans are implemented comprehensively and efficiently, that estimate may be plausible.

Meanwhile, $5 billion will be used to expand public transit, while $15 billion more will go toward maintenance and operations. A program expanding options for bicyclists, walkers, and others not using cars will get $2 billion. All in all, 38 percent of the regional planning budget is devoted to getting people out of their cars and onto buses and trains. Transit ridership, the commission boldly asserts, will increase "72 percent between now and 2030."

That conclusion is hard to swallow. Transit isn't fulfilling its promise in Atlanta now, and the trends in the city's census data aren't much different from what's happening in Minneapolis. Atlanta's regional work force is 2 million. Transit ridership increased to 75,272 workers in 2000 (an 8 percent increase), hardly making a dent in general commute patterns. And despite that modest uptick in absolute numbers, transit's market share fell from 4.7 percent of all commuting trips in 1990 to 3.7 percent in 2000.

Almost 2 million jobs will be added to the regional economy during the next 30 years. If Atlanta achieves its transit ridership goal, 129,467 people will be using mass transit in 2030. And even then, transit's commute share would fall. Put another way, Atlanta is investing almost 40 percent of its transportation budget on less than 4 percent of the market, and the latter number is shrinking.

This might be a worthy investment if the main beneficiaries were people too poor to afford cars or otherwise restricted from getting around. But the commission wants to compete with the automobile-to get working-class and middle-income commuters out of their cars and onto buses and trains.

Even our more skeptical analysis of the city's transit trends might be overly optimistic. According to the U.S. Department of Transportation's National Transit Database, Atlanta's bus transit system logged 235 million passenger miles in 2003. That's down from 273 million miles reported in 2000. Atlanta's subway system reported 487 million passenger miles in 2003, down from 504 million reported in 2000. So transit use is falling even in absolute terms. For transit to turn around and increase market share would be unprecedented.

Atlanta's policy makers can still shift course. Traffic in the city has become so bad that in 2004 Gov. Sonny Perdue convened a task force that called for making congestion reduction the top priority for regional transportation planners. The Atlanta Regional Commission, along with the Georgia Department of Transportation and other agencies, recently agreed to set specific targets for reducing congestion in absolute terms, as measured by travel delay in peak periods, by 2030. Local officials appear receptive, but have not yet revealed how and to what extent they will follow the recommendations.

We'll see if the follow-through lives up to the promises. If it does, we can only hope the rest of the country's urban planners are paying attention. The myths that have held Atlanta back are hardly unique to that city.

City House, Country House

In 2005 the Urban Transportation Monitor, a biweekly industry newsletter, surveyed more than 600 transportation professionals to find out their thoughts on traffic congestion. About 19 percent responded. Of those, 45 percent thought the profession was "doing all it can do" to stop congestion. Half thought congestion was the result of too many people using their cars, and 45 percent attributed it primarily to the desire to live in low-density suburbs.

The preferred solutions were predictable: 51 percent thought mass transit should be improved or expanded, and 50 percent thought the government should manage demand better by getting people to telecommute or carpool. Only 29 percent believed increased highway capacity could be a cost-effective way to reduce congestion significantly. (The survey did not ask whether new capacity should be provided if it were privately funded.)

Many believed the problem is simply too many cars. Fifty-one percent said one of "the main reasons for the high level of congestion in many metropolitan areas" is the desire "of many to use cars for all their trips." Indeed, of the 11 options offered by the survey, that was the biggest vote getter. For traffic engineers, planners, and other transportation professionals, the solution to traffic jam is to keep us from using our automobiles.

The planning profession clings tenaciously to its foundational myths. Even as overwhelming evidence to the contrary piles up, planners keep claiming that cars are inefficient and socially destructive; that expanding road capacity isn't practical; and, most fundamentally, that the government can determine how we choose to travel by planning where and how we live.

That last assumption is the logical conclusion of a rather sophisticated (if largely incorrect) way of looking at human behavior. It's rooted in a common-sense observation: How we live influences how we travel. If we live on a farm, we are going to travel by car. Buses simply don't go out to farms to pick people up and take them into town for work or to buy groceries. Trains don't either. A neighbor might, but she would probably be driving a car and doing this as a service because you don't have a car. School buses are the exception that proves the rule. They pick up a large number of kids, but only because they're being delivered to one destination, the school building.

The flip side is the experience of the Manhattanite. If someone lives in the densest neighborhood of an American city, cars are costly, frustrating, and inefficient. Most Manhattan residents can get to their destination far more efficiently using the subway, taking a bus, or walking. Because parking is so costly, they also can get around fairly efficiently using taxis.

So people in dense urban areas have more choices, and personal automobiles are inefficient ways to get around town. Congestion, in fact, leads people to use alternative modes of transportation. Many regional planners, like those in Atlanta, conclude that the way a region develops dictates how people are likely to travel and what transportation strategies are most feasible. And the way to influence development patterns, they believe, is to carefully plan where and how much to invest in the transportation system. But proximity to work is only one of many factors people consider when finding a home; other criteria, such as price, neighborhood safety, and proximity to good schools, are often deemed more important than living close to the office.

Of course, Atlanta is not Manhattan. In fact, it's virtually the opposite. At 1,783 people per square mile, Atlanta is the poster child for low-density residential development. The New York metropolitan area is three times as dense, with 5,309 people per square mile. Manhattan's density is even higher: more than 50,000 people per square mile.

According to the Atlanta commission, "Land use is an important determinant of how people choose to travel. No other variable impacts [mobility] to a greater extent. The Regional Development Plan policies help shape future growth and protect existing stable areas by encouraging appropriate land use, transportation, and environmental decisions."

To say this is an exaggeration would be charitable. While land use can influence travel behavior in small and crude ways, to claim that it is the biggest factor distorts the mainstream research on the subject. A 2004 study sponsored by the Federal Transit Administration (FTA) cautioned against the tendency to "overemphasize vertically mixed uses such as ground-floor retail and upper-level residential." In particular, it noted that "outside of dense urban locations, building mixed-use products in today's marketplace can be a complex and risky proposition; few believe that being near a train station fundamentally changes this market reality."

This isn't to say that these developments can't generate more transit riders. The FTA study found that those living near rail stations were five to six times more likely to commute using transit than other residents. While those seem like dramatic effects, the majority of commuters near transit stations (often two-thirds or more) still use cars to get to work. Moreover, many of the people living in these transit areas were transit users already. They just moved so they could be closer to transit.

Put differently, if 5 percent of a region commutes using transit-about the national average-then 25 or 30 percent of those living in a transit-oriented development will commute using transit. This is consistent with case studies of transit use in San Francisco and Chicago. (Incidentally, those results invariably come from studies of predominantly heavy rail commuter systems, such as subways. Light rail and buses are more fashionable in planning circles these days, but they're also slower and carry fewer riders.)

To get such high use rates, densities have to be very high. The traditional American home with a private yard doesn't fit this model. The typical new house in the United States is built on about one-fifth of an acre. A study in San Francisco found that doubling densities from 10 units per acre to 20 units per acre would increase transit's commute share from 20 percent to 24 percent.

In short, even cramming four times more people into the typical U.S. subdivision of 4-5 units per acre would produce only a modest uptick in transit use. And it isn't an uptick for the region. It's an uptick for the neighborhood-those living within a quarter mile of a transit stop. There is virtually no effect beyond the immediate vicinity of the transit stop, regardless of density.

At these densities, Americans would literally have to give up any hope of having a decent-sized yard and most would have to live in townhouses. The land use pattern would have to fundamentally change, resembling the landscape more common in the carless 19th century than in the highly mobile and adaptable 21st century.

Forget, at least for the moment, whether the government should effect such a sweeping change. It almost certainly can't. In a forthcoming report, Adrian Moore of the Reason Foundation (the nonprofit organization that publishes this magazine) and Randal O'Toole of the Thoreau Institute examine data from the National Personal Transportation Survey and find that doubling an urban area's density would, at most, reduce the total number of car trips by 10 percent to 20 percent. No U.S. urban area has managed to double its density or to reduce car travel by such magnitudes.

Real Solutions

Believe it or not, there are ways to reduce traffic congestion, even if most politicians and planners haven't been eager to adopt them. Here are five potent suggestions, ideally done not alone but in conjunction with one another:

Creative construction. Expanding capacity doesn't always mean adding lanes to congested roads, although that's often a good idea as well. In densely populated Southern California, portions of the highway network are elevated well above the ground, including the Harbor Transitway approaching downtown Los Angeles. In Texas, San Antonio and Austin have double-decker freeways as well. In 2006 Tampa opened its cross-town expressway, an elevated road built in the median of an existing four-lane highway.

If going up is a problem, you can also go down. Australia has done an effective job of using tunnels to connect highways while preserving neighborhoods, an excellent alternative to destroying businesses and homes.

Smarter management. Building new capacity can get you only so far. The Federal Highway Administration estimates that half of all congestion could be eliminated simply through better management of the existing road network. Among other approaches, this could mean metering freeway ramps, turning two-way streets into one-way streets, and improving traffic light coordination. According to the Institute of Transportation Engineers, better-coordinated lights can reduce stops by as much as 40 percent, thereby cutting gas consumption, emissions, and travel times.

Market pricing for roads. One especially fruitful idea is high-occupancy toll (HOT) lanes, which allow drivers who put the highest priority on quick commutes to pay a premium for uncongested lanes. These have been built in Denver, Houston, and-yes-Minneapolis, among other cities. In Atlanta several private companies have submitted plans to build new HOT lanes on their own dime. During rush hour, the congestion difference between the special lanes and the regular lanes can be the difference between going 15 miles per hour and doing 65.

Areas with lots of car pool lanes could convert those to HOT lanes, add some connectors, and create a congestion-free HOT network. Transit boosters, take note: It would be easy to tweak the arrangement to guarantee bus riders a speedy trip too.

Market pricing for parking. On 99 percent of our trips we park for free, thanks largely to the minimum parking requirements embedded in our zoning codes. Eliminating those requirements would allow market forces to reflect the true cost of parking. Instead of adhering to arbitrary regulations that often order more spaces than necessary, developers would have greater flexibility to build only the number of spaces that is needed. Workplaces would be more likely to adopt parking cash-out programs, which give employees who do not drive to work a share of the money that otherwise would have gone toward parking costs. Employees would be more likely to work from home.

Market pricing for parking would reduce traffic too. If drivers had to pay the full cost of parking, they might be less inclined to take certain trips, thus putting a dent in congestion. More important, when parking is scarce but free (or underpriced), drivers have an incentive to keep the spots as long as possible. When it is scarce but costs money, drivers are less likely to dally. One additional result: Other drivers have less need to circle around and around, hoping eventually to spot an empty space.

Traditional parking meters can be notoriously inconvenient, but they aren't the only way to pay for parking. Aspen, Colorado, uses a variety of new technologies, including personal in-vehicle meters. The town determines its parking rates by zones; prices are highest in the city center and drop the further you are from the core. Motorists simply park, type in the number of their parking zone, turn on the meter, and hang it from the rearview meter. A timer deducts the prepaid amount until the driver returns. No one has to hunt for loose change.

Privatization. We're much more likely to adopt ideas like the above when roads are built and managed by companies responding to market incentives, not by government officials responding to planning fads and political clout. Private companies can create and operate highways using toll revenues as a funding source. The government can also convert existing roads to privately managed systems to allow improvements and expansions of the existing network.

For a spectacularly successful example, consider the 407 Electronic Tollway outside Toronto. This innovative road isn't fully private, but it was built by a private company (the Canadian Highways International Corporation) and is now managed by another private company (407 International) that bought a 99-year lease from the government of Ontario. Yet another company, Hughes Electronics, equipped it with an electronic toll-collecting system that eliminates toll booths and the congestion they can cause.

Baby, You Can Drive Your Car

There is a fundamental disconnect between transportation planners and the typical American commuter. Most travelers believe the car is a good thing, a source of freedom and mobility. Giving up the flexibility of the private automobile reduces our quality of life; it's a step back, not a step forward. That's the main reason the use of mass transit is declining in the U.S., despite the billions of dollars poured annually into such systems.

Yet transportation planners believe public transit and sharing rides with strangers increases the typical American's quality of life. It doesn't, and our behavior reflects this. That's why the vast majority of us choose not to use public transit.

Back in Minneapolis, Sue may hop aboard the Hiawatha Line from time to time. But when even well-off, condo-dwelling rail fans like her continue to rely on their cars, the currently dominant school of transportation policy seems destined to create many more traffic jams than transit users.

Samuel Staley is Research Fellow

Ted Balaker is Producer






;