They called Atlanta “the poster child of sprawl.” In the 1980s and 1990s, the metropolitan region was growing in leaps and bounds, unimpeded by natural barriers that could stop the seemingly boundless suburban development driven by an influx of refugees from colder climes. “Atlanta may be the fastest-spreading settlement in human history,” said national real estate expert Christopher Leinberger in the mid-1990s, when the metro Atlanta commute shed swelled to 50 miles in every direction, and farther on the popular north side.
Fast-forward to 2014, six years after the collapse of the housing bubble, and Atlanta has become a poster child for the opposite trend: infill development, or “walkable urbanism” as Leinberger calls it. “In this real estate cycle, 60 percent of development in Atlanta has happened on less than 1 percent of existing land area,” he said, basing his figures on an analysis he and associates at George Washington University conducted. “That’s compared to the 1990s, when the overwhelming share happened in a conventional suburban locale.”
Walkable locations — neighborhoods where a mix of homes, offices and shops exist in close proximity to each other — are growing more than twice as fast as the market as a whole. Development at the fringe of the metro area has remained at almost a dead stop, he said.
While exemplary for the radical nature of the shift, Atlanta is hardly alone, Leinberger and other experts said. Washington, D.C., Chicago, Boston, San Francisco and Seattle all of which in the 1980s and 1990s saw rapid development at the fringe and slow or no growth in the core have reversed that trend in recent years. While an up-and-coming region like Nashville may be flirting with a sprawl revival, it is the exception and not the rule, even among the Sunbelt cities known for sprawling in the past.
Of course, Atlanta and dozens of other regions are still dealing with the legacy of spread-out development. In April, metro Atlanta topped a ranking of most-sprawling large metros in a report from researchers at the University of Utah and Smart Growth America. The researchers used an index based on residential and employment density; the degree to which offices, homes and shops are segregated or integrated; the share of the region’s jobs and residents clustered in centers; and whether the street network is walkable as well as drivable or only drivable.
Researchers then looked at “how sprawl affected the household pocketbook and individuals,” said Reid Ewing, the Utah planning professor who led the research. Comparing the combined cost of housing and transportation in more sprawling and less sprawling regions, “We found that the increased transportation costs associated with the automobile, combined with the lower housing costs in sprawling areas, was equal to the increased housing costs in more compact areas where transportation costs were low,” Ewing said. “Transit and walking is much higher in the compact places, and drive times are shorter. People are spending less of their time and money commuting.”
In terms of safety on the roads: “We looked at crash rates and found people in compact places have half the fatality rates. There are more crashes in compact areas, but they aren’t nearly as likely to be fatal. We speculated that is because they are at much lower speeds.” Individual health: “People in sprawling areas are far more likely to be obese and suffer from illness associated with it, including high blood pressure, coronary heart disease and type II diabetes — controlling for race, income and education. People in compact places are more active and spend less time in the sedentary activity of driving.” And in terms of economic mobility: “You’re much more likely to move up the economic ladder in a compact place than a sprawling place. We speculated that it is mostly a result of job accessibility. In a more compact place low-income people have better access to jobs.”
But while Ewing’s sprawl study found significant remaining impacts from decades of sprawling development, it also certified the slowing of the trend. Even as metro areas have continued to add people — accounting for 85.4 percent of the population according to a March census report — sprawl was “about 1 percent worse” from 2000 to 2010, Ewing said. “That’s not a lot,” he added. “If we had looked at 1990 to 2000 there would have been a much greater difference.”
“The sprawl energy is dimming, there’s no doubt,” said Robert Lang, a professor of sociology at the University of Nevada – Las Vegas specializing in metropolitan growth and development. “I was a skeptic for a long time, but now I’m convinced.”
The evidence shows up in a number of trends, Lang and others said.
For decades after World War II the rate of driving — the number of miles driven per American each day — was increasing faster than population. The typical household was driving more and more each year, and commercial travel was going farther. Women joined the workforce in larger numbers, the economy grew and people had the means to travel more.
But by the 1990s, the miles driven per household plateaued. Since the early 2000s, the miles driven per person “has been flat or dropping,” said Eric Sundquist, managing director of the State Smart Transportation Initiative at the University of Wisconsin. “The economy only explains part of it,” he added. Driving dropped significantly when gas prices soared and the economy soured in 2008, “but it hasn’t really rebounded.”
He cites a number of factors: “We built a lot of highway infrastructure, but we’ve gone almost as far as we can. After a time, the market for cars got saturated. Baby boomers are aging out of the peak driving years, from ages 20 to 50.” At the same time, commute distances might have outgrown people’s tolerance. “Trees can’t grow to the sky and we can’t drive 25 hours a day. People are making other choices with their dollars and hours. So now instead of driving to qualify (for a home loan), you take money out of your travel budget and put it in your house in a more convenient location.”
The drop in driving seems likely to have some staying power, Sundquist said. Members of the millennial generation — those now ages 18 to 34 — are driving almost a third less than the baby boomers did at their age. They are delaying getting their drivers licenses and buying cars. Some of this is attributable to their high levels of school debt in an era of poor employment prospects. But even those who are working are choosing more walkable locations and opting to use their smart phones rather than drive, he noted.
Since the real estate bubble burst more than six years ago, the housing market has undergone convulsive change and is not likely to return to post-war “normal,” said John McIlwain, senior fellow for housing at the Urban Land Institute. In that time, the homeownership rate has dropped from 69 percent to below 64 percent, and is likely to dip further, he said. Post-war sprawl was driven by middle class buyers looking for larger houses on larger lots. But today middle-class buying power is shrinking, even as production costs have risen, McIlwain said.
“For this decade, at least, suburban outer sprawl is over,” McIlwain said. “The boomers moved to the suburbs 20-plus years ago, they have done that. If they move, they will not go to a further suburb, and many are looking at more urban locations.” Their replacement buyers, Generation X, form a smaller cohort, “a historical aberration,” McIlwain said.
The millennials are as large a generation as the boomers, but their housing future is a huge question mark. “They all say they want to own their own home, but the question is when,” said McIlwain. “The first-time market is severely constrained by tighter credit.” Debt burdened and facing dim employment prospects, many are unable to buy a home. At the same time, “By and large they say they want more urban environments, and hope to be in an urban-style town center when they have kids, even if it’s in a more suburban location. This is the first generation since we had automobiles that has less interest in owning cars and driving them. This is not the generation that will renew the outflow to the outer ring.”
If and when the millennials start buying, “There is a lot of land in existing areas to develop before you go out further. There are a lot of failed suburban development projects, and much of the land has negative value now. No one will finance the big master-planned communities on large pieces of land.”
The age of sprawl also was the age of the mall, said Lang. “Malls were a rejection of the traditional urban product,” open-air Main Streets with shops at the ground floor and offices or apartments above.
Today, “There are no traditional covered malls under construction,” he said. The replacement trend of earlier this century was the “lifestyle center, an attempt to ape the urban.” Now those are pretty much done, as well, he said, “as people have noticed that actual urban is better.”
Covered malls, with their shifting anchor stores, themes of the moment and ephemeral boutiques, were always expected to need periodic refreshing. “But today the re-do is to tear it down and re-build as mixed use with parking hidden away,” Lang said. At the same time, an excess of suburban retail real estate is intersecting with the rise of Internet shopping to suppress demand for new retail space.
Developer capacity to meet shifting market demand
Last decade’s housing bubble obscured a trend that may also have had a hand in deepening the effects of its collapse, McIwain said. In the aftermath of the collapse, “The closer in your property was, the closer to transit, the higher the values. Those areas recovered much more quickly,” he said. Today, as development is picking up, “more is taking place in infill, by repurposing shopping centers and pre-existing buildings. We’re seeing smaller houses, smaller lots, to bring cost down in close-in locations.”
These conditions reveal an underlying shift in market demand that existed before the bubble burst, but that the development industry lacked the capacity to meet until the last decade or so, Lang said. “When I was at Fannie Mae in the late 1990s, we needed to think about ways to finance development in urban places that had not seen investment in years and had no market comparables. We needed to convince people to finance mixed use, we needed more people who could design it. It seemed impossible. But now we know how to do these things again. It’s now well established and well regarded. … You’d have to unlearn a serious amount of knowledge for things to go back to where they were.”
Urban-style development — whether in downtown cores or close-in suburbs — is occurring not just in regions with a long history of it, such as the Northeast, but in Sunbelt regions like Dallas, Charlotte, Phoenix, Salt Lake City and Orlando. “These are all cities in conservative places that are succeeding with light rail and walkable neighborhoods,” Lang said.
The role of policy and design movements
“In the 1950s,” McIlwain said, “the white middle class wanted to leave the crowded, dirty cities, and the Levitts showed you could make a lot of money” by selling homes in suburbs in developments such as their Levittowns. “Cities are now clean, vibrant, great places to live so there’s a draw to them, not a repulsion from them.”
Federal policies of subsidizing highways and home loans gave the 1950s trend a shot of adrenaline, while local governments adopted single-use zoning at an accelerating clip. “Policy clearly helped drive the suburban migration and the shape that suburbia took,” Sundquist said. For the last several years, he added, federal and local policies affecting development and transportation have been changing, and in some places they are driving change. “But in most cases the policies are still on the books and we have to do battle with them to get them to allow what the market demands now.”
Lang gives some credit to the movements for smart growth and new urbanism for resurrecting the policy and design intelligence that shaped walkable urbanism before the automobile, and updating it for the modern era. “There are big forces at play, but the millennials wouldn’t have anything to live in if people hadn’t figured out how to do urban development again. They can do it because people my age and older were thinking about it in advance.”
In 1967, President Johnson rolled the population clock to 200 million, and “the next 39 years until we hit 300 million was almost all sprawl,” he recalled. “The next 100 million, the share that will be built up in the existing space is going to increase sharply. Our first century was about laying down a network of cities, the next period was building around them, and this one is about building within.”