Removing Freeways - Restoring Cities

Produced by the
Preservation Institute

Home Page

Tear It Down!
by John Norquist

Portland, OR:
Harbor Drive

San Francisco, CA:
Embarcadero Freeway

San Francisco, CA:
Central Freeway

Milwaukee, WI:
Park East Freeway

Toronto, Ontario: Gardiner Expressway

New York, NY:
West Side Highway

Niagara Falls, NY:
Robert Moses Parkway

Paris, France:
Pompidou Expressway

Seoul, South Korea
Cheonggye Freeway

Other Freeway Removals

Freeway Removal
Plans and Proposals

From Induced Demand
to Reduced Demand
by Charles Siegel



From Induced Demand
To Reduced Demand
by Charles Siegel

The first generation of freeways is approaching the end of its lifespan.

New York's West Side Highway began construction during 1920s and crumbled during the 1970s.  The freeways built during the freeway building boom of the 1950s and 1960s have begun to obsolesce: many already need rebuilding and many more will need rebuilding during the next few decades. In general, the cost of rebuilding them will be more expensive than new freeway construction.  As John Norquist has said, "governments around the country can't afford to rebuild the highway infrastructure without bankrupting their economy." 

As freeways obsolesce, there will be political battles to decide what to do with each one - to rebuild it or to remove it. 

Removing Freeways

By now, it should be clear that removal is the best alternative - and the only alternative that helps us deal with looming environmental problems such as global warming. We have seen that traffic engineers were wrong when to predict that freeway removals would lead to gridlock. We have seen that even in cities where there was strong political opposition, freeway removals turned out to be successful and popular after they were completed.

When we look at freeway spurs, parts of larger freeway plans that were never completed, it is obvious that removal is best.  There is no significant impact on capacity of the total freeway network, and there are obvious benefits, because tearing down the freeway reclaims land for new development or parks and helps revive adjacent neighborhoods.

When these freeways run through downtowns, there are huge economic benefits to tearing them down.  For example, we have seen that:

  • Milwaukee spent $25 million to demolish the 1-mile-long Park East freeway, while it would have cost $100 million to rebuild that 30-year-old freeway.  Removing the freeway opened 26 acres of land for new development, including the freeway right of way and surface parking lots around it, which have already attracted over $300 investment in new development, in addition to stimulating development in surrounding areas.
  • San Francisco increased nearby property values by 300 percent by tearing down the Embarcadero Freeway and opening up the waterfront was opened up, stimulating the development of entire new neighborhoods.

Of course, it is more radical to tear down mainline freeways rather than just freeway spurs, because this reduces capacity on the entire freeway system. Nevertheless, cities are beginning to remove mainline freeways: 

  • Niagara Falls is removing the Robert Moses Parkway in order to slow people down and encourage them not to drive as far.  Just as building this parkway encouraged tourists to take longer trips and drive right through to Niagara Falls, Canada, removing this parkway is meant to encourage tourists to take shorter trips and stop in Niagara Falls, New York.
  • Paris is considering closing the Pompidou Expressway as one element of a larger plan to reduce automobile use by reclaiming land from the automobile.  It is also converting traffic lanes on major streets to bus lanes, as part of the same plan.
  • Seoul has removed the Cheonggye freeway and restored the river that it covered in order to stimulate the economic revival of central Seoul's Dongdaemun district.  It has built busways to replace the freeway capacity, and it the goal of this plan is to reduce automobile use from 27.5 percent to 12 percent of all trips.

In the 21st century, it should become common to tear down mainline freeways to reduce automobile dependency, because this will be necessary to deal with global warming and other environmental challenges of the coming century..

Induced Demand

It should be cleaar by now that it was a mistake to build urban freeways. Because they blighted the older neighborhoods that they sliced through and made it easier to commute from remote new suburbs, the freeways encouraged suburban sprawl - so they generated more traffic, which quickly filled the freeways beyond capacity. Freeways that were supposed to handle projected demand for decades became congested in just a few years, because of the traffic that they themselves generated.

This is what transportation planners call "induced demand." Building freeways encourages people to drive longer distances: in the short run, people begin to drive to regional malls rather than local stores, and in the longer run, they move to lower density neighborhoods where they have to drive further for all their trips.

One study found that, within five years after a major freeway is built in California, 95 percent of the new road capacity  fills up with traffic that would not have existed if the freeway had not been built.1 Other studies in different places show different levels of induced demand, but they generally agree that, within a few years, more than half of the new capacity fills with traffic that would not have existed if the road had not been built.   

In Great Britain, where there is a very active anti-freeway movement, transportation planners are no longer allowed to count reduced travel time as a benefit of building a new freeway. The Department of Transport has adopted a guidance document saying that cost-benefit studies on new freeways must assume that elasticity of demand may be as high as 1.0 with respect to speed - which means that average trip length increases as much as speed increases, so building freeways and increasing speeds just lengthen trips and does not save any time.2     

When post-war American freeways generated sprawl and longer trips, transportation planners began to theorize that the average person budgets a constant amount of time for transportation, so that higher speeds just make people travel longer distances.  This idea was first advanced by Yacov Zahavi of the U.S.  Department of Transportation, who studied changed in travel patterns between 1958 and 1970 and found that people did not spend any less time traveling, though all the freeways built during that period let people travel faster.3   

Follow-up research confirmed his conclusions.  It showed that the amount of time that Americans spend commuting to work has remained constant since the 1840s, when the move to the suburbs began as a reaction against the industrial revolution, though there have been vast changes in technology since then.4  The total amount of time that Americans budget to transportation also tends to remain constant, about 1.1 hours per day.5   As speeds have increased, suburbs have sprawled over more land, the malls have gotten bigger, and people have driven further to get to their jobs or go

Per Capita Vehicle Miles Traveled in the United States

The distance that the average American travels has doubled since the 1960s.
source: Census Bureau, Statitsical Abstract of the United States

Reduced Demand

If people have a constant time budget for transportation, we should expect that tearing down freeways and reducing speeds will reduce the distances that people travel.  Just as building freeways and increasing speeds causes induced demand, removing freeways should cause reduced demand.   This is sometimes called "traffic evaporation."

Studies have shown that reducing road capacity does reduce traffic - but not as dramatically as increasing capacity increases traffic.

In 1998, a comprehensive study of the issue was published.  With funding from the city of London and British government, a team of researchers at University College, London, examined 60 cases where road capacity was taken away from cars, with examples from the UK, Germany, Switzerland, Italy, The Netherlands, Sweden, Norway, the USA, Canada, Tasmania and Japan.  Often, there were predictions that reducing capacity would cause gridlock, but the researchers found that, though there was sometimes short-term disruption, there were no cases of long-term gridlock.  Most drivers moved to parallel streets or changed their travel time to avoid congestion.  Some drivers changed their mode of travel, changed where they carried out activities, or changed where they lived to avoid congestion.  Overall, 14 to 25 percent of the traffic that had used the removed capacity simply disappeared, on the average.6

In some cases, reductions were more dramatic than this average.  In 1973, when New York city's West Side Highway collapsed, 53 percent of the traffic that had used that freeway simply disappeared.7  Presumably, traffic was reduced by such a large amount because people in the center of New York city have good transit alternatives and shopping within walking distance of their homes.

If only about 20 percent of traffic disappears on the average when we reduce road capacity, it is best to taking a more comprehensive approach to reducing traffic at the same time that we remove mainline freeways. 

Two policies that can reduce traffic immediately, at the same time that freeways are removed, are:

  • Parking Cash-Out: Businesses could be required to give employees commute allowances instead of free parking.  Employees could use the allowance to pay for the parking they used to get for free, could use it to pay for transit, could keep part of the allowance if they car-pooled to work, or could keep the entire allowance if they walked or bicycled to work. It is estimated that this policy could reduce commuter traffic (and peak demand for road space) by about 20% immediately and by even more as better transit service is provided.
  • Congestion Pricing: As in London and Stockholm, drivers could be charged a fee for driving into the central business district at times when roads are congested.  The revenues could be used to pay for better public transportation.  This policy has been very successful where it has been tried, and the fee can be set at the level needed to reduce congestion to a manageable level.  San Francisco is currently studying congestion pricing for its central business district.

Rebuilding Our Cities

In the long term, we need to rebuild our cities to make them less auto-dependent by promoting:

  • Public Transportation: Though states have had some flexibility to spend federal funding on either highways or transit since 1992, they still spend far more on new freeway capacity than on public transit: between 1992 and 2001, states spend 29% of federal highway funds on new freeway capacity and only 5% of these funds on public transit (with the bulk of the funding going to maintain existing roads).8  At the same time that we reduce freeway capacity, we need to provide transit alternatives - particularly transit on exclusive rights-of-way, which gives people a way to avoid traffic congestion.
  • Smart Zoning and Smart Growth:  Today, most zoning requires developers to build at suburban densities, which are so low that they cannot support decent transit service or local shopping, so people have no choice but to drive.  The new urbanists have developed zoning codes that allow developers to build more compact neighborhoods of single-family houses, with transit and shopping streets within walking distance.  We need smart growth policies to these walkable neighborhood of single homes and denser neighborhoods of apartment buildings around transit nodes and corridors. In this way, we would gradually rebuild our freeway-oriented cities as transit- and pedestrian-oriented cities.

It is plausible that building freeways has such a strong effect in inducing traffic precisely because it has been supported by other public policies. Zoning has required low densities, separate land uses, and abundant parking. Suburbanization was also jump-started by FHA loans, which initially were available only to buyers of only low-density suburban housing.  Businesses virtually all provide free parking for customers and employees who drive, and nothing for people who use public transpotration or walk.  These have been the policies during the era of freeway building, and they help to explain why freeways generated so much traffic. 

Likewise, it is plausible that removing freeways has less effect in reducing traffic because it is not supported by other  public policies.  To make freeway removal more effective, we should zone to allow more compact neighborhoods, we should give federal loan guarantees to apartment buildings and mixed-use projects with housing above shopping as well as to single-family houses, and we should shift funding from new freeway capacity to new transit systems. 

Beyond the Conventional Wisdom

Today, the conventional wisdom among environmentalists focuses on two of the three things that we must do to rebuild our cities: shifting funding from new freeways to public transportation and zoning for pedestrian and transit-oriented development.  But the conventional wisdom shies away from the third thing that we must do: reducing freeway capacity and slowing traffic. 

Apparently, we still intimidated by the traffic engineers who say that reducing capacity will cause gridlock, though they have been proven wrong by the recent British study of cases where capacity actually has been reduced.

Instead, the conventional wisdom seems to be that we should underground mainline freeways when they obsolesce, so they do less damage to the surrounding neighborhood.  Yet undergrounding freeways is tremendously expensive.  The "big dig" in Boston cost $14.6 billion to underground just two miles of freeways - far more than the initial projection of $4 billiion.

The usual argument for undergrounding is that we cannot simply eliminate existing mainline freeways, because removing them would cause congestion and drivers would waste many hours in traffic.  But this is no different from the traffic engineers' argument that building new freeways will save drivers time.  In reality, because people have a constant time budget that they devote to transportation, they will eventually change their patterns of activity to accommodate lower speeds.  We can deal with congestion in the short term using parking cash-out, congestion pricing, and other forms of transportation demand management. And we can make it possible for people to change their patterns of activity in the long term by promoting public transportation and smart growth.

The conventional wisdom - that we should build pubilc transportation and pedestrian and transit-oriented development - is not enough in itself.. As long as people can travel at 60 or 70 miles per hour on the freeways, they will drive to do their shopping at big-box stores and drive to work at jobs in Edge Cities, just as often as they shop at the stores near their homes and take transit to jobs in city centers. 

This is what has happened in the past: in the 1930s, American cities were all pedestrian and transit oriented, but much of this transit and local shopping has vanished because of competetion from freeways and freeway-oriented shopping.

The same thing continues to happen today. New York City has plenty of local shopping, but it loses sales tax revenues to suburban counties because people drive there to shop at big-box stores. To recover these tax revenues, it is under pressure to allow more big-box shopping within city limits, though these new big-box stores would obviously draw even more people from neighborhood shopping streets.

To build environmentally sustainable cities, in addition to promoting public transportation and smart growth, we need to remove mainline freeways and, in most cases, to replace them with boulevards.  Boulevards can carry a substantial amount of traffic - but at a lower speed than freeways.  Boulevards allow cross-traffic, so they do not slice apart neighborhoods like freeways. In fact, with a service and parking lane on each side, separated from the main traffic flow by landscaped medians, they can provide an attraction that helps pull the neighborhood together.  Boulevards can also include light rail in a separate right of way. 

The idea of removing mainline freeways sounds radical today, but it will prove to be necessary during the twenty-first century, in order to deal with global warming and the depletion of fossil fuels. 

After all, the changes that we need in the twenty-first century are no more radical than the changes that occured during the twentieth century, when we transformed our traditional pedestrian- and transit-oriented cities into automobile-oriented sprawl by building freeways and by zoning for low-density suburban style development.  During the twenty-first century, we need to transform the sprawl back into something more like traditional cities by building public transportation, promoting smart growth, and removing freeways.

The twentieth century's freeways and suburban zoning gave us more traffic, gave us an ugly landscape of strip malls and sprawl, gave us neighborhoods where it is impossible to go anywhere without driving, and are now giving us global warming.

By contrast, freeway removal, transit, and smart growth would give us less traffic, would give us neighborhoods that are attractive enough that we would enjoy walking, and would leave a healthier global environment to our children and grandchildren.



1. Mark Hansen and Yuanlin Huang, "Road Supply and Traffic in Californian Urban Areas." Transportation Research A, Volume 31, No 3, 1997, pp. 205-218.

2. UK Department of Transport, Report of the Standing Committee on Trunk Road Assessment, December 1994 and Guidance document 24 L, based on this report.

3. Yacov Zahavi, Travel Over Time, Report PL-79-004 (FHWA, U.S. Department of Transportation, 1979).  Yacov Zahavi and Antti Talvitie, "Regularities in Travel Time and Money Expenditures," Transportation Research Record 750 (TRB, National Research Council, Washington, D.C. 1980) pp. 13-19.  Yacov Zahavi and J. M. Ryan, "Stability of Travel Components Over Time," Transportation Research Record 750 (TRB, National Research Council, Washington, D.C., 1980) pp. 19-26.  A later study updated Zahavi's analysis using data through 1990 and concluded that he was right to say that people have a constant time budget that they devote to traveling: Gary Barnes and Gary Davis, Land Use and Travel Choices in the Twin Cities, 1958-1990, Report No. 6 in the series Transportation and Regional Growth (Minneapolis: Center for Transportation Studies, 2001)

4. J. M. McLynn and Spielberg, "Procedures for Demand Forecasting Subject to Household Budget Constraints" in Directions to Improve Travel Demand Forecasting: Conference Summary and White Papers, HHP-22 (Washington DC, Federal Highway Administration, 1978) pp. 115-197.

5. J. M. Ryan and B. D. Spear, "Directions toward the Better Understanding of Transportation and Urban Structure," in ibid., pp. 199-247.

6. Cairns, Hass-Klau and Goodwin, Traffic Impact of Highway Capacity Reductions: Assessment of the Evidence (Landor Publishing, London, 1998)

7. Jill Kruse, "Remove It and They Will Disappear" Progress, published by Surface Transportation Policy Project, March 1998.  

8. Table 2 (insert) in Progress, published by Surface Transportation Policy Project, February, 2003.


copyright 2007 by Charles Siegel

return to top