Skip to content

Lewyn Addresses America

A little politics, a little urbanism- I also blog 100 percent on urbanism at and

The Strong Towns website is sponsoring “No New Roads” week; the website will contain a variety of articles etc. on transportation issues related to that theme.  So I thought I would revive something I wrote on the topic some years ago (if you want to look at the footnotes download this article)

If state and federal policy caused our urban crisis, the logical solution is to stop the policies that led to the crisis. Because highway spending has been a significant cause of suburban sprawl, [FN472] we can take a significant bite out of both sprawl and big government by eliminating sprawl-generating highway spending. Specifically, state and federal governments should prohibit the use of their funds to build or widen roads in newer suburban areas. Because highway spending totaled $101 billion in 1997, [FN473] such a ‘paving moratorium’ would give taxpayers a significant break (including, ironically, drivers, whose fuel taxes pay for more than half of government highway spending). [FN474] A paving moratorium would not prevent settlement in existing suburbs–but would prevent government from creating new suburbs by building more highways, and would thus prevent government from turning today’s suburbs into deserted slums. Government justifies new and widened roads on the ground that more roads, not fewer, are needed to deal with traffic congestion. [FN475] Butthe*367 claim that new roads eliminate congestion is at best speculative. Admittedly, if new and widened roads did not affect development patterns, a new or widened road might reduce traffic congestion. But in reality, highway building affects where people live and work. If government builds highway X to suburb Y, homeowners and businesses will soon move to subdivisions near X’s interchanges, thereby increasing traffic along the interchanges. [FN476] Thus, ‘[b] uilding more highways to reduce traffic congestion is an exercise in futility. Whenever it is done, more people take to their cars, and before long the roads are as clogged as ever.’ [FN477] Even people and groups sometimes identified as pro-sprawl admit as much. As Joel Garreau of the Washington Post has written, ‘[t]he more capacity you add, the more likely you are to make the place more popular . . . creating more traffic.’ [FN478] Mr. Garreau is hardly an anti-sprawl activist; for example, he has described the status quo as the ‘manifest pattern of millions of individual American desires over seventy-five years.’ [FN479] Similarly, the National Association of Home Builders (which advocates accelerated road construction) [FN480] conducted a survey that reveals that highway access would influence 55% of respondents to move to a new community–more than any other amenity. [FN481] By admitting that highways encourage movement to areaswith *368 highway access, the NAHB has effectively conceded that highways shift development to suburbs (thus making those suburbs more rather than less congested). [FN482] Numerous studies suggest that ‘induced traffic’ eliminates some or all of the reduction of congestion caused by new roads and road widenings. For example, Mark Hansen, a professor of transportation engineering at the University of California at Berkeley, used statewide California statistics in concluding that new road capacity is almost entirely offset by induced traffic within five years. [FN483] A study conducted by Robert B. Noland, a former transportation analyst at the Environmental Protection Agency, similarly found that a 10% expansion in roads produced a 2.8% rise in travel over two to four years. [FN484] These traffic increases arise because in the short run, motorists switch from other routes, because they abandon mass transit and drive instead, and because development may shift people and jobs to areas near the highway. [FN485] In fact, studies such as Hansen’s, if accurate, may actually overestimate the benefits of new roads by failing to account for the medium- and long-run changes in development plans caused by new and widened roads (that is, the changes that occur more than four or five years after the road is built or widened). For example, in 1991, Montgomery County, Maryland (a suburb of Washington) widened Interstate 270 to as many as twelve lanes to reduce traffic congestion. [FN486] According to Sidney Kramer, Montgomery County executive from 1986 to 1990, ‘[y]ou saw a tide of development go forward because of that improvement.’ [FN487] One of the high-growth suburbs created by the I-270 widening, Germantown, Maryland, grew from just over 41,000 people in 1980 to 70,000 people in 1998. [FN488] In turn, the growth of Germantown and nearby suburbs caused traffic to increase. In fact, traffic along I-270 has surpassed the levels statehighway *369 planners forecast for 2010 in their 1984 study of the proposed widening. [FN489] The Maryland highway department reported that the ‘1997 volume at Route 28 in Rockville was 193,000 vehicles [per] day– 2,000 more than the 2010 projection.’ [FN490] According to David Palank, an area real estate broker, ‘[w]ith all the lanes that are there, it just doesn’t seem to be moving that quickly . . . I haven’t found any relief at any time. It seems like it was congested and continues to be congested.’ [FN491] If I-270 was an aberration, areas that increased road space would have experienced a reduction in congestion during the 1990s, or at least less congestion than areas that did less road-building. But recent studies show otherwise. The Hartford, Ct., and Providence, R.I., areas experienced similar population growth between 1982 and 1997. [FN492] But Hartford’s road capacity stagnated, while Providence increased its road mileage by 59%. [FN493] If road-building reduced congestion, Providence would have far less congestion than Hartford. But a study by the Texas Transportation Institute (the official research agency for the Texas Department of Transportation and the Texas Railroad Commission) [FN494] revealed that the two areas had similar levels of traffic growth and traffic congestion. In 1997, the cost of congestion per eligible driver was $390 in Hartford and $360 in Providence (Nos. 49 and 50 of 68 areas surveyed). [FN495] Rush-hour congestion increased by 200% from Hartford and 225% in Providence between 1982 and 1997. [FN496] Annual delay per driver increased by 283% in Hartford and 320% in Providence between 1982 and 1997. [FN497] In other words, Providence built far more roads, yet congestion increased in Providence just as rapidly as in Hartford. The correlation between free-flowing traffic and free-spending road builders is equally weak in fast-growing metro areas. For example, Charlotte and Fresno had comparable population growth rates (64% and 57%). [FN498] But Charlotte increased its highway mileage by 113hile *370 Fresno’s road-building lagged behind its population growth (with only a 27% increase). [FN499] Charlotte’s congestion cost $680 per driver, while Fresno’s cost only $315. [FN500] Annual delay per driver increased by 356% in Charlotte and only 171% in Fresno, [FN501] while peak hour congestion increased by about the same amount in both areas (229% in Charlotte and 225% in Fresno). [FN502] Ironically, drivers are sometimes the biggest losers from road-building: When states favor road-building over routine street maintenance, roads become rutted and packed with potholes. A recent survey by The Road Information Program, a group financed by the road construction industry, shows that 35% of roads in Detroit and New Orleans are in poor condition. [FN503] Over 30% of roads were in poor condition in three other metro areas (Los Angeles, Indianapolis, and San Jose), and 20% to 30% of roads were poor in fourteen others (San Francisco, Chicago, Baltimore, Sacramento, Grand Rapids, Norfolk/Virginia Beach, Oklahoma City, Denver, Dallas, Houston, New York, Washington, Philadelphia, and Austin). [FN504] By an odd coincidence, all of these areas increased highway capacity in the 1980s and 1990s. [FN505] For example, Detroit’s highway mileage increased by 21% (far ahead of its anemic 5% population growth) while New Orleans’s highway mileage increased by 45% (despite that region’s 4% population growth). [FN506] It, therefore, appears that some states are letting existingroads *371 deteriorate so that they can build new roads instead. In sum, both common sense and actual experience support the view that suburban road-building creates sprawl without mitigating congestion. Thus, continued road widening and roadbuilding is pointless, if not harmful.


%d bloggers like this: