"Roman Candle" turns 20: Secrets of Elliott Smith's accidental masterpiece (slideshow)
Elliott and the friends with whom he recorded in middle school in Texas (photo courtesy of Dan Pickering)
Despite the struggles of the 1970s, or perhaps because of them, sprawl moved on. It spread over wider territories. It mutated into new forms. The eye was assaulted by landscapes never seen before. Fields of McMansions sprang up in the countryside, gated communities cowered behind stucco walls, office towers were sprinkled among parking lots.
The outward wave was now propelled by its own momentum. New homes and workplaces, reachable only in a car, dumped traffic back into older neighborhoods. There the fight against change was fought even more fiercely, as the urge to wall out the automotive flood reinforced the sentiment of status-seeking. Development was driven out onto the fringe; highways, widened to carry ever more traffic, became unwalkable; sprawl begot more sprawl.
Planners had come to reject the old orthodoxies. But the new theories, when they made their way into rules at all, were heavily watered down, and what got built seemed little better than what came before. Advertised cures like growth controls and cluster zoning all too often made things worse. Sprawl seemed immune to all attack. It grew like a cancer, ever changing, leaping over whatever obstacles were set in its path, swelling even as the city that gave birth to it shriveled.
By the 1980s the interstate highway system planned in 1956 was nearly finished, aside from canceled urban segments. Yet this gargantuan feat of engineering had hardly banished traffic jams. Housing tracts spread outward along the new interstates, their inhabitants compelled to drive farther and more often. Cars sat interminably on crowded downtown expressways, and backups spread to once-empty suburban streets. More roads, it could be seen, only brought more traffic.
Congestion worsened, but highway construction went on. Interstates got wider and new expressways appeared on maps. Road builders, closely allied with the oil, automobile, and trucking industries and powerful suburban real estate interests, had enormous clout. New roads brought new sprawl and even more driving. The average American drove two-and-a-half times as far in 2000 as in 1960.
The highway lobby did face resistance. The urban freeway revolt had faded, but environmentalists remained on the attack. The Clean Air Act of 1970 empowered the Environmental Protection Agency to block new highways that threatened to pollute the air. Still, even after the law was strengthened, state highway agencies found ways to get around the rules. They claimed, for example, that road widenings would purify the air by getting rid of stop-and-go traffic. They built carpool lanes as a way to clean up. These arguments, however dubious, won acceptance from public agencies reluctant to confront the well-connected road interests.
The new roads had to be paid for, and gasoline taxes rose steadily. The highway lobby, initially opposed to the gas tax, reversed course and went along with a jump to 3 cents a gallon in 1956 to pay for the interstates. After President John F. Kennedy hiked the tax another penny in 1961, increases were avoided for years. But the inflation of the late seventies drove up the cost of new highways, and repair bills mounted as the interstates aged. States were raising their tax rates, and in December 1982 Congress approved an increase to 9 cents in a federal transportation bill proposed by President Ronald Reagan.
The federal gas tax was raised twice more in the early 1990s, but it has not gone up since. In the states, too, increases have lost popularity. Between 1998 and 2008, more than half the states did not raise their gas tax at all, and only five states increased it faster than the rate of inflation.3 Other trends squeezed revenues as well. As cars got better gasoline mileage, drivers paid less tax per mile. And the spread of suburban superblocks forced traffic onto major arteries and off the local streets built by city and county governments. The indirect subsidy of state highways by real estate taxes shrank, because state highway departments derived a smaller portion of their tax revenue from gasoline consumed on roadways they did not pay to maintain.
As suburbs spread and highways could no longer punch through open countryside, the cost of new roads exploded. Price tags now reach two or three billion dollars for roads that serve just one sector of a metropolis. By the 2000s, highway budgets were in a tight pinch.
The highway lobby was no more deterred by lack of money than by air pollution or its failure to reduce congestion. Money went into new roads, while old ones were allowed to fall apart. States issued bonds to be paid back with future federal aid. New toll roads, rare in the years when federal funds were pouring in, once again made their appearance.
The road builders had largely given up on the centers of big cities, but they still wanted to lay asphalt in built-up suburbs. Here they faced a thorny problem when they tried to justify their plans. When traffic backs up, people drive less than they would if the roads were clear. New lanes added to the highways quickly fill up with trips that no longer have to be forgone, and cars move no faster than before. Spending money on highway construction seems a waste if it doesn’t make traffic move faster.
To maintain the promise of congestion relief, highway backers came up with the idea of adding toll lanes next to toll-free expressways. Tolls in these lanes vary from hour to hour. As traffic thickens, the rate goes up so that more drivers will be unable to pay. If you can still meet the price, you get to drive the speed limit.
These toll lanes were quickly dubbed Lexus lanes, and they deserve the name. A study showed that drivers with incomes above $100,000 were four times more likely than those who earn less than $40,000 to have used the toll lanes on their last trip. Tolls can reach levels that seem astronomical to drivers accustomed to free interstates, yet they rarely bring in enough money to pay back the cost of construction. Most Lexus lanes need heavy subsidies.
Highways are thus segregated by economic class, much like suburban neighborhoods. Lexus lanes, by design, serve a minority—if most of the cars were in the pay lanes, the free lanes would move at the speed limit and there would be no reason to pay. The tolls are primarily an allocation mechanism, and only incidentally a source of revenue. Their purpose is to deter those less able to pay from using the new lanes. Those wealthy enough to afford the tolls bypass the traffic jams, while everyone backed up on the free lanes gets to pay the bills.
Not only did the interstates and their clones proliferate, but lesser roadways were dedicated ever more exclusively to automotive use. Those who still tried to walk found little welcome. Children’s games were no longer tolerated, even where dead ends and closed loops ensured that traffic would not be blocked. Some states banned street play by law, and there are suburbs where even play on sidewalks is forbidden. In 2000, the Federal Highway Administration announced that “children at play” signs would no longer be allowed—because they might encourage play.
For a new suburban generation, the only way to get around was in a car. In 1969, 49% of elementary school pupils still walked or biked to school; by 2009 the figure was 13%. Instead of walking to friends’ houses, children were chauffeured to prearranged play dates. A Rockville, Maryland, mother who allowed her ten-year-old to take the public bus to school risked an investigation for child abuse. Other parents complained to the principal that her behavior was dangerous—in a place where car keys are routinely handed over to seventeen-year-olds.
The layout of new suburbs and redeveloped downtowns reinforced the primacy of the automobile. Wide lanes encourage cars to speed. Pedestrians, forbidden to cross on one side of an intersection, need three green lights to get to the other side. If they reach a corner when the light is green, they face “don’t walk” signals timed to pen them in so cars can turn faster. At one suburban intersection—in an area designated for transit-oriented development—crossing the street is a journey of 81⁄2 minutes across twenty-eight traffic lanes.
The curved roadways that Lewis Mumford and George F. Babbitt had so insisted on were now ubiquitous. Even residential streets that carried little traffic widened—room is needed, it is argued, for fire engines to turn around.
There was a subtle but profound alteration in the way street corners are built. Curbs no longer meet at right angles; they swing around in broad curves. It became standard even in cities for the curb to start bending back 25 feet from the cross street. On busy suburban roads, the bend begins even farther from the corner. Those on foot must choose between dangerous crossings of broad asphalt expanses and annoying zigzags to where the road narrows. Cars round the turn at highway speed. The simple act of walking down the street is so perilous that pedestrians are sometimes warned to wear reflective clothing, as if they were in the woods during hunting season.
These changes were no mere whim of car-loving traffic engineers. Behind them stood the lobbying might of the trucking industry.
The truckers had fought for decades to put bigger vehicles on the roads, but they were long stymied by the railroads. A major battleground was Pennsylvania, where the Pennsylvania Railroad held sway over the legislature and limits on trucks were especially strict. A few weeks before the 1950 election, the Pennsylvania Motor Truck Association divided $76,000 between the chairpersons of the state Democratic and Republican parties. It was, the association’s treasurer later conceded under oath, like betting on both teams at a baseball game, but he countered that “nothing was hidden, it was all out in the open.”
The truckers gained ground in the 1970s as their old antagonists weakened. But they still faced strenuous opposition from local governments and the American Automobile Association. Even highway engineers objected; they worried that bridges weren’t built to carry the weight of big trucks. Just before the 1974 election, the Truck Operators Nonpartisan Committee made last-minute campaign contributions to 117 congressional candidates from both parties. Six weeks later, the House of Representatives reversed an earlier vote, and weight limits were raised on interstate highways.
In December 1982, the truckers won full victory. The Reagan administration agreed to their demands in exchange for the industry’s acceptance of a tax increase that hit trucks harder than autos. Weight limits were raised again, and state limits on the length and width of trucks were overruled. Tractor-trailers could have trailers up to 48 feet long; soon the limit in most places was 53 feet.
A key provision, not fully understood by critics when the law was rushed through a lame-duck Congress, legalized the big trucks on many local roads as well as on the interstates. Road-builders had a new justification for designs that encourage cars to speed; pedestrians, ignored when the issue was under debate, were the victims. Lanes grew wider; curbs were pushed back at intersections so that extra-long vehicles could make the turn. And, because it was written into the statute, the neighbors had no way to object.
As the years went by, builders put up fewer apartments. From 1969 to 1973, apartments were 38% of all new housing. The ratio dropped sharply when the bubble burst in 1974, and it never recovered. A smaller multifamily housing boom came in the mid-eighties, but even at its peak apartments did not exceed 32% of new homes, and in the post-2000 housing bubble they were only one out of every six units. In 1972 900,000 apartments went up; in the best years after 2000 the number barely reached one-third of that.
Why did builders switch from apartments back to one-family houses after moving so eagerly the other way in the sixties? There was no lack of demand. Economic trends, demographics, and shifting tastes all favor apartments. Single-family houses are less affordable; since 1973 real income has gone down for people in the bottom half of the income distribution, who make up the majority of apartment dwellers. There is no greater demand for yards for children’s play; nearly half of all households had children in 1960 but only a third in 2000. And gentrifying neighborhoods, with their row houses and apartment renovations, show that multifamily living is hardly losing popularity.
Only the tightening of land use regulation in the nimby era can explain the falloff in construction of apartment houses. Their builders face stricter zoning, growth controls, and aroused neighbors.
A telltale sign that regulation is at work is the near-disappearance of the two-to-four–family house. If dislike of urban living were causing the drop in apartment construction, two-family houses would stay in fashion—they are much more like single-family homes than high-rises are. But when rules are tightened, small rental properties cannot bear the cost of lawyers and paperwork. Census data show that the construction of two-to-four–unit dwellings has fallen off even more than that of larger buildings. In 1972 more than 140,000 new dwelling units of this kind came onto the market; in the best year after 2000, the number is 43,000.
The latest housing market collapse has hit small multifamily buildings especially hard. In 2010, just 11,400 apartment units of this type were built—only 5000 or so new two-to-four–family structures in the entire United States. Yet there are willing buyers and able builders for such homes. Even after the bust, gentrifiers snap up this kind of house in dilapidated neighborhoods. The small construction firms that did mansionizations during the boom would have no trouble building two-family houses. The demand is there; only the law stops them.
Homeowner associations, once rare outside the wealthy neighborhoods that could bear their operating costs, moved downscale after the FHA issued its endorsement in 1963. Some localities forced developers to set up associations that would maintain streets, sewers, and parks. The current residents wanted more tax revenue without the cost of maintaining new infrastructure. Only five hundred subdivisions had homeowner associations in 1962; in 1975 the figure rose to twenty thousand; and by 1994 80% of all new housing was subject to these private governments. Architectural review, imposed only by a few high-end developers in the early part of the century, became a routine feature of suburban life. Covenants, never expiring and nearly impossible to change, embalmed neighborhoods with a permanence that the developers and zoners of the 1920s could only aspire to. Constitutional protections of free speech and free assembly vanished when streets and parks were privately owned.
The privatization of the suburban public realm advanced another step with the gated community. By the end of the century, fenced-off subdivisions were ubiquitous in suburbs of the south and west, their walls sometimes lining both sides of a highway. The design impulse behind the superblock was taken to an extreme, with entry limited to a few openings where uninvited visitors could be turned away. Inside the fence, the homeowners association was more powerful than ever, controlling physical access as well as land use and design. And the automobile held residents in a tight grip; with so few gates open, travel by any other means was a near impossibility.
The sales pitch for these subdivisions appealed to emotions of fear and snobbery. The biggest market was retirees. They had lived the suburban migration of the 1950s and watched, usually at a distance, the urban troubles of the 1960s. Retirement communities let no one under fifty-five reside within the gates; those compelled by the vicissitudes of life to raise their grandchildren must move out. Another target group was parents of young children, who had grown up in a tightly controlled world of tract houses and play dates. Observers of gated communities noted the unusual homogeneity of their population, not just economically but racially as well.
Walls and fences provided little real protection against lawbreakers—the crime rate in ungated neighborhoods of similar nature was already low, and barriers did little to lower it. But a vague dread of the different and unfamiliar took hold. Such fear was behind the much-debated death of seventeen-year-old Trayvon Martin in 2012. Martin’s killer, George Zimmerman, patrolling an economically hard-pressed gated subdivision in Sanford, Florida, may or may not have been guided by racial animus. But there is no dispute about Zimmerman’s first explanation of what made him call the police: “This guy looks like he’s up to no good or he’s on drugs or something. It’s raining and he’s just walking around looking about.” The gated community was so thoroughly suburban that “just walking around looking about” could be seen as a criminal act.
Workplaces moved steadily out from the city center. Factories had been leaving the cities since the 1940s, eager to avoid city-based unions and in search of more room for spread-out assembly lines. Offices now followed, with corporate headquarters joining high-tech businesses in the flight to suburban office parks. By 1998, only 23% of the jobs in the country’s hundred largest metropolitan areas were within 3 miles of the center.
What was happening here? Jobs began moving into the suburbs somewhat later than the spread of residences; it is tempting to think that the purpose of the exodus was to shorten commutes. But the jobs did not go to the same places as the homes. Population spread out fairly evenly around the cities; employment clustered where the housing was highest-priced, in what real estate experts call the “favored quarter.” Big shopping malls moved in the same direction; department stores were attracted to people who had spending power.
The most important factor determining where a corporate office moved was proximity to the chief executive’s home and golf club.18 Jobs concentrated near the fanciest houses. In places like Stamford, Connecticut, Mountain View, California, and Bethesda, Maryland, old town centers near rail stations blossomed into urban downtowns. More often, “edge cities” grew up on leftover land near freeways, vast agglomerations of office towers and shopping malls set among parking lots. Tysons Corner, the biggest edge city, borders the expensive Virginia suburbs of McLean and Great Falls; Houston’s Post Oak is next to River Oaks; Los Angeles’ Century City is across the road from Beverly Hills.
Edge cities forced nearly everyone into cars. They were so dense that cars were unavoidably stuck in traffic, but nearly impossible to reach by any other means. Buses ran, at best, once or twice an hour. They took long detours in and out of superblocks, following routes so twisted that they sometimes looked like they needed tomato sauce. Few but those too poor to own an automobile would put up with the delays.
Even within the superblocks, six- and eight-lane roads ran between the parking lots, nearly impassible on foot. To cross the street to buy lunch, you got in your car. The CEO had a short drive home on back roads and an assistant who ran errands during the day. The lower ranks sat in traffic jams.
It was the market that jammed jobs into a favored quarter, but the consumer was not sovereign in this market. The consumers of office space are the people who work there—and they had no say. Their bosses forced them into long commutes on overcrowded highways. Indeed, when jobs moved to the suburbs to escape unions, the employers’ very purpose was to deprive workers of a voice in decisions.
As these trends came together, house-building accelerated. Restrictions drove up the price of land, and a self-reinforcing dynamic set in. Houses came to be seen as a financial investment as much as an object of consumption. Rising prices no longer curbed purchasers’ demand for dwelling space, as conventional economic theory holds, but stimulated more of it. Financial bubbles emerged in the housing market as prices in many cities soared in the late 1970s and again a decade later. Both booms were cut short by economic recession, but by the late 1990s prices were shooting upward even faster than before.
By now suburbs were no escape from urban ills. Tract housing aged and did not always age well. Traffic often was worse than downtown. Poverty and crime plagued run-down garden apartments.
With unions largely banished, suburbs abounded in low-wage service and construction jobs, and immigrants flocked to them. The working poor, hardly able to afford a car even when they qualified for a driver’s license, led a precarious life amid the landscape of strip malls and subdivisions. From a rented room, a dangerous walk along the side of the highway led to a long wait for the bus.
Affordable rentals were scarce, and homeowners struggled to pay their mortgages. Yet it was illegal to divide houses into apartments. In earlier generations, idle construction workers could make up for lost wages by adding rental apartments onto small houses. Now this escape route was shut off. Declining neighborhoods, still zoned single-family long after their veneer of prestige had worn off, were unable to upgrade by the residents’ own efforts. They were dragged down by rules meant to uphold their social standing.
Elsewhere, the old formula for escape was still in use. Workplaces leapfrogged to the outer edge of the favored quarter; real estate was cheap, and the boss’s drive to work would be against traffic. People moved outward too; the population was whiter, the houses newer, the roads not yet filled up. Exurban subdivisions sought to stand out from the Levittowns, and the average size of a newly built house ballooned from 1785 square feet in 1974 to 2582 in 2008. Large lots and thick foliage created an illusion of rural isolation. Homes were harder to walk to, they were farther from the city, and they were more restricted than ever.
Middle-income buyers, their incomes stagnant, needed generous financing to pay for the big houses, and the pace of building went up and down with changes in the lending market. Savings and loan associations, earlier the main source of private mortgage loans, could no longer meet the demand; they had sought quick profits in commercial real estate after a 1980 deregulation allowed them to diversify and were crushed by the bursting of the speculative bubble that followed. Fannie Mae and Freddie Mac, federal agencies that had been privatized in the sixties, wanted to keep growing, and they filled the gap. They bought up mortgages with money borrowed cheaply thanks to taxpayer backup. The mortgages, in a process called securitization, were packaged into bonds and sold.
Fannie and Freddie could push things only so far. Still somewhat tied to the government, they had rules that deterred overly risky lending. To really blow up the housing market, private enterprise had to work its magic. Wall Street did not shy from this challenge. It took securitization a step further, slicing and dicing loans and sorting the pieces into bonds that were said to be much safer, thanks to the magic of statistics, than the original mortgages.
Repeal of the Glass-Steagall Act in 1999 deregulated banks, allowing mortgage lenders to merge with stockbrokers. New financial giants arose, seeking growth at all costs. The faking of paperwork became routine; lenders and homebuilders had sales targets to meet. Bonuses were paid on what you did this quarter, and, when disaster struck, banks that were too big to fail would be bailed out anyway. The flood of new capital fueled a vast construction boom. By the time it peaked, unscrupulous bankers were encouraging bad loans so they could place bets against them.
Before the bubble ended in disaster, it gave one more push to suburban sprawl. There were few bankers left with the local knowledge to judge the value of a small parcel in a city neighborhood. Wall Street had squeezed them out. A big downtown office tower could still borrow on its own merits, but lesser mortgages had to be assembled into packages and sent out into the financial marketplace.
The rationale for this process rested on large numbers—whether an individual mortgage will be paid is unpredictable, but if bond buyers took the average of enough loans and compared them to similar loans that have already paid off, they could know what to expect. So the securitizers needed mass-produced mortgages, resembling as much as possible the mortgages of years past. To write mortgages that all looked just the same, the big banks wanted little boxes—and not-so-little ones—that all looked just the same. Developers were happy to supply the copycat buildings that eager lenders demanded. Strip malls and office parks lined the highways, all built in the same pattern. South Carolina, seen from the road, could have been California. Stores, offices, and houses were financed by Wall Street number-crunchers who cared only for cookie-cutter buildings that fit the established car-centered mold. The bankers’ bonuses rose and fell with their deal-making, so how easily a loan could be packaged mattered more than the value of the property behind it.
Fueled by speculation and financial fraud, the wave of building gathered force and grew to a mighty crest. In the twelve months of 2005, 1.7 million one-family houses were built, a quarter-million more than any year before or since. The houses of the new outer suburbs were bigger and more spread out, and their inhabitants had no choice but to drive more.
In 2008 the flood began to retract. The real estate bubble, already letting out air for three years, collapsed so suddenly that the world’s financial system nearly went down with it. Barely a quarter as many houses were built in 2009 as four years before. Prices tumbled, falling most in the new subdivisions farthest from downtown. Some developments were abandoned, half-finished, when builders went bankrupt. Others were awash in foreclosures and empty houses.
The ever-bigger houses of the boom years had been matched with ever-stricter covenants and ever-tighter zoning. Architecture and governance alike were designed for perpetual affluence, and the aftermath of a popped bubble brought troubles no one had prepared for. Soon vigilante patrols were mounted to keep squatters out of empty McMansions. Mosquito control officers sprayed abandoned swimming pools; Lee County, Florida, had so many foreclosures that it used helicopters to find the pools. A gated community outside Atlanta, in the habit of fining residents $250 for planting unauthorized flowers, discovered that it lacked means to shut down a bordello operating in one of its houses.
The great migration was over. A half-decade after the crash, suburban building has resumed, but it will never again be the same inexorable outward surge. The landscape created by the migration remains, embodied in houses, stores, roads, habits, and laws. What we do with that landscape will determine what kind of country we make in the century to come.
Excerpted from “Dead End: Suburban Sprawl and the Rebirth of American Urbanism” by Benjamin Ross. Copyright © 2014 by Benjamin Ross. Reprinted by arrangement with Oxford University Press, a division of Oxford University. All rights reserved.
Elliott and the friends with whom he recorded in middle school in Texas (photo courtesy of Dan Pickering)
Heatmiser publicity shot (L-R: Tony Lash, Brandt Peterson, Neil Gust, Elliott Smith) (photo courtesy of JJ Gonson photography)
Elliott and JJ Gonson (photo courtesy of JJ Gonson photography)
"Stray" 7-inch, Cavity Search Records (photo courtesy of JJ Gonson photography)
Elliott's Hampshire College ID photo, 1987
Elliott with "Le Domino," the guitar he used on "Roman Candle" (courtesy of JJ Gonson photography)
Full "Roman Candle" record cover (courtesy of JJ Gonson photography)
Elliott goofing off in Portland (courtesy of JJ Gonson photography)
Heatmiser (L-R: Elliott Smith, Neil Gust, Tony Lash, Brandt Peterson)(courtesy of JJ Gonson photography)
The Greenhouse Sleeve -- Cassette sleeve from Murder of Crows release, 1988, with first appearance of Condor Avenue (photo courtesy of Glynnis Fawkes)