top of page




In his 1940 book, Magic Motorways, industrial and futuristic designer Norman Bel Geddes said, "It would seem, though, that American traffic experience has shown that when major routes go directly into cities, they cause congestion and confusion.  When on the other hand, they avoid cities on the Bureau's by-passing "beltline distribution roads," they are forced out of their way, and as a result they lose a quality that is essential to a highway, namely that of being the shortest distance between two points.  Actually, there is a third alternative which makes the problems connected with both of these solutions unnecessary.  It is to consider highways as straight-line routes laid out on a direct course between the environs of cities, instead of directly from the center of one city to the center of another.  Tradition, true enough, calls upon the road to steer straight for the heart of town.  But if the purpose of the motorway as now conceived is that of being a high-speed non-stop thoroughfare, the motorway would only bungle that job if it got tangled up with a city.  It would lose its integrity.  The motorway should serve heavily populated areas, but it does not have to connect population hubs directly.  A great motorway has no business cutting a wide swath right through a town or city and destroying the values there; it’s place is in the country where there is ample room for it and where its landscaping is designed to harmonize with the land around it.” [1]


In agreement was Joseph Barnett, Chief of the Urban Roads Division of the U.S. Public Roads Administration.  In March 1946 Proceedings of the American Society of Civil Engineers, Express Highway Planning in Metropolitan Areas, Barnett wrote, “To dump the [freeway] traffic onto downtown surface streets already overcrowded would defeat a prime purpose of free-flowing facilities – the relief of traffic congestion.”  In his book, Downtown, Its Rise and Fall, 1880-1959, Robert Gogelson states, “Geddes and Barnett were in a minority, at odds not only with most traffic experts but also with most downtown businessmen and most local, state and federal officials.”  According to Gogelson, Bel Geddes

did advocate for “feeder roads” allowing constant speeds of 50 mph that connected to inner-city “express boulevards” with traffic at similar speeds.  Both of these road types are essentially expressways in of themselves.  He did provide the caveat that travelers would generally transfer to public transportation for entry into cities. [2]


In his book, The Highway and the City, Lewis Mumford said, “The purpose of transportation is to bring people and goods where they are needed, and to concentrate the greatest variety of goods and people within a limited area, in order to widen the possibility of choice without making it necessary to travel.  A good transportation system minimizes unnecessary transportation; and in any event, it offers change of speed and mode to fit a diversity of human purposes.”  [3]  Mumford also said, in his book Urban Prospect, “Arterial roads, ideally speaking, should engirdle the metropolitan area and define where its greenbelt begins; and since American cities are still too impoverished and too improvident to acquire greenbelts, they should be planned to go through the zone where relatively high-density building gives way to low-density building….It is on relatively cheap land, on the edge of the city, that we should be building parking areas and garages:  with free parking privileges, to tempt the commuter to leave his car and finish his daily journey on the public transportation system.” [4]


Below is a 1929 sketch by planner Benton Mac Kaye of the “Townless Highway” and the “Highwayless Town” from Dartmouth College Archives. Noting the disorder created by highway-oriented development, Mac Kaye said in Zone the State Highways: The Lesson of the Mohawk Trail (1928), “This is not the architecture of the well-ordered town; it is the architecture of the slum: not the slum of poverty but the slum of commerce.”  In 1930 he subsequently advocated a preferred approach in The Townless Highway, “A town is worthy of a personality…To have a personality, however, a town must first of all be a separate geographic unit.  It must have open space around it.”  He further said that this is counter to commercial strips which “…create ‘roadtowns’ instead of a group of buildings around a common center or purpose.  Such development does not meet the conception of a true town or unit of society; it is not a town, it is merely a linear spreading of humanity.”

In his 1994 book, Rural By Design, Randall Arendt provides the example of Route 6 in Eastham, Massachusetts, to demonstrate the disaster that can occur from unrestricted zoning along highways.  During a large portion of the 1980’s, this highway strip of numerous roadside commercial establishments had the highest fatality rate in the Commonwealth.  “The continued, almost miraculous expansion of roadside commercial strips right through the lean years of the Great Depression so impressed tax-hungry local officials that they referred to these strips as ‘miracle miles,’ a name by which they are known today in some quarters.” [5]





Gilles Duranton and Matthew Turner, in  The Fundamental Law of Road Congestion:  Evidence from U.S. Cities, cite economists Gilles Duranton and Matthew Turner who have found that vehicle miles traveled tend to increase as much as one-to-one with the number of miles of new highway.  The authors conclude that expansion of either highways or public transportation is unlikely to have much impact in relieving traffic congestion.  Studies by the Texas Transportation Institute and other entities are well-documented in demonstrating the persistence of road congestion over time despite the expansion of transportation facilities. [6]


In their book Sustainability and Cities, Peter Newman and Jeffrey Kenworthy document their analysis of 1990 per capita kilometers (PCKM) and wealth or per capita gross regional product (PCGRP) data for 37 global cities or very large urban areas.  The analysis shows that mobility is not clearly related to wealth.  Cities with high GRP, mainly European and wealthy Asian areas, tend to have high wealth and low mobility.  The U.S. cities have 7.3 times the car use of wealthy Asian cities but only 1.26 times the level of GRP.  The U.S. cities have 2.41 times the car use of the average European city but only 0.85 the level of GRP per capita.


In the graph below, Newman and Kenworthy’s data is expressed through a scatterplot which converts PCKM to per capita vehicle miles traveled (PCVMT).  A polynomial trend line is used to obtain the highest correlation.  As can be seen PCGRP tends to rise rapidly with small incremental increases in PCVMT but starts leveling-off at around 3,000 to 4,000 PCVMT.  Caution should be used as this is not a relatively large sample and does not control for other variables.  But it is consistent with other research that the marginal benefits of increased travel are generally not worth the additional costs beyond a certain point.  What that point is exactly is debatable and likely varies by a certain extent depending on the geographic area.  But it is consistent with the example of a household converting from two vehicles to one which showed a decrease of PCVMT from about 9,600 to 5,700.  [7]



There is value in coordinating land use and transportation in larger urbanized areas with policies that reduce VMT.  Planners often use one or more of three tools to measure the value of a proposed transportation investment:  economic impact analysis; cost-effectiveness analysis; and benefit-cost analysis (BCA).  Another approach is the use of empirical documentation of past project impacts to assist planners in using the information for analyzing new projects.


Economic impact analysis looks at the effects a project or policy has on the local, regional or national economy and the associated economic relationships. Complex input-output models are often used to look at the economic transactions among industries to estimate changes or multiplier effects in employment, income and GDP.


Cost-effectiveness analysis compares the costs of different alternatives to achieve a specific level of benefit.  An example is the relative cost of options that reduce accidents in an intersection or corridor to a certain level. The option that reaches an established target level at the lowest cost would be the most cost-effective.


Benefit-cost analysis (BCA) compares the benefits and costs of a project over an extended period than can include the useful life of a completed project for up to 30 years and more.  Its use is a best planning practice as it measures the extent in which overall society is better-off due to a project.  This holds true, particularly, if it conducts a fully comprehensive/exhaustive accounting of all benefits and costs.  It is unique in that it monetizes benefits (or disbenefits) of impacts on:  the ecosystem, noise, different types of emissions (e.g. carbon dioxide, particulate matter, and nitrogen oxide), accidents (injuries, deaths, property damage), and residual value (value of infrastructure at then end of the analysis period).


BCA is an excellent way for evaluating projects and alternatives against each other to determine the one with the best benefit-cost ratio.  A number of metropolitan planning organizations (MPOs) use BCA, or modified versions, to evaluate major capital projects that expand capacity.  However, many do not use BCA which makes it difficult to understand if a particular project is worth implementing.  Note that jobs created from a project are generally not a benefit.  Jobs are an opportunity cost, rolled into project expenses, that represents the loss in value of work that could be done more productively to generate higher wealth.  However, increasing productivity can be counted as a benefit.


The downside of BCA is that:

  • It does not address equity.

  • Additionally, there can be disagreement on the appropriate discount rate used to account for the changes in the value of money over time (or if one should be used at all).  A range from 3% to 7% is typical.  The U.S. DOT TIGER program requires all capital project applications to include a BCA using a 7% rate which makes it more difficult for projects to achieve positive B/C ratios.

  • Perhaps the biggest concern with BCA is the lack of consensus on the criteria that should be used.  It is generally agreed that time savings is a critical element that should be used to evaluate all transportation projects. In project alternatives analysis, transit options often may not be able to compete with highway options as the latter can tend to have significantly higher time savings.  This can be exemplified when the value of ecosystem/crop production losses or resource consumption costs (depletion of non-renewable resources, environmental/human health costs of production, military costs to protect energy interests, etc.), among others are not accounted for in BCA.


An excellent resource for comprehensive all-inclusive BCA is the Victoria Transport Policy Institute (


Much of the interstate highway system has or is reaching its useful life.  Stakeholders considering alternatives for roadway reconstruction often neglect to conduct robust evaluation of non-conventional options such as full/partial demolition or capacity reduction.  This is because these appear counterintuitive to rationally addressing transportation problems of congestion and growing population/employment.  Additionally, job creation is always an important indicator.  Consequently, project sponsors will often focus on cost-effectiveness analysis or economic impact analysis to quantify employment, income and GDP outcomes directly related to the project.  Indirect outcomes of these metrics could be higher in a capacity-reduction scenario that is often never considered.   Further, BCA is often neglected due to either the aforementioned problems or because the B/C ratio may not be substantially positive.


An example of the dilemma presented by the often contrasting outcomes of EIA and BCA is with the Illiana Corridor Expressway Project (Illiana Project).  This proposed new 47-mile four-lane rural limited access expressway would connect I-65 in northwest Indiana with I-55 in northeast Illinois just south of the Chicago, IL-IN UZA.  The project purportedly would relieve congestion on parallel route I-80 about 15 miles to the north.  In February 2014, an EIA was conducted on the Illiana project for the State of Indiana by the Economic Development Research Group, Inc.  (EDRG).  Findings show that , “Overall, from 2015 to 2048, the project is expected to generate an average of 3,378 jobs per year and a cumulative total of $6.85 billion in personal income and $21.3 billion in economic output (in constant 2012 dollars) in the regional economy.”  This includes alleged transportation efficiency improvements based predominantly on changes to vehicle miles traveled (VMT) and vehicle hours traveled (VHT) pursuant to travel demand modeling.  The annual fiscal impact (i.e. net tax revenues) from new population and employment is estimated at about $40.7 million.  The study concludes that the Illiana Project will have a positive impact on the Chicago region.


BCA conducted by R. Arkell, AICP in 2014, Illiana Corridor Expressway Benefit-Cost Analysis, found that VHT reductions are indeed substantive for both heavy trucks and light vehicles.  VMT reductions for heavy trucks are also significant, however, much of this is offset by increases for light vehicles.  Changes in emissions are relatively inconsequential.  The BCA also determined that there would be net benefits from accident reductions in terms of reduced fatalities, injuries, and property damage.  However, the BCA found costs due to ecological value losses from land used for the ROW itself, induced development, and the loss in revenues from farming operations.  Operating and maintenance costs for the Illiana Project are a moderate disbenefit.  The overall Illiana Project cost of about $1.5 billion is the largest barrier to feasibility of the project.  The BCA concludes that at a mid-range discount rate of 5% the Illiana Project has a benefit/cost (B/C) ratio of 0.95 and net present value of -$98,406,513.  Generally, B/C ratios of less than 1 or not substantially above 1 are good indicators that a project is not worth the cost, i.e. society is not better off.


Public transportation ridership performance can be compromised, even in dense urban areas, when expressways penetrate to and through urban cores to facilitate personal vehicle mobility.  Perhaps most importantly, is that noise, vibration, emissions and other externalities within and adjacent to urban expressways create an inhospitable environment counter to quality of life expectations.  Consequently, BCA can reveal potential opportunities to realize multiple planning objectives of reducing PCVMT, increasing transit usage/modal balance, and reducing negative externalities to the environment and nearby residents and businesses.


In the Chicago urbanized area, Chicago Transit Authority (CTA) trains operate within the median of three separate expressways, a legacy of the interstate construction program beginning in the 1950s.  At the time of their development, shared right-of-way (ROW) appeared to be a rational solution to eliminate conflicts with street traffic while reducing maintenance costs of elevated rail systems.  The design is still attractive today as exhibited by other cities locating light rail adjacent to or within expressway ROWs.


The downside of this type of design is that transit-oriented development (TOD) is difficult, if not impossible, due to the constant drone of traffic and other associated negative externalities from expressways.  Additionally, the traffic noise experienced by transit commuters both at stations and on trains is excessive compared to the relative quietness experienced by Metra commuter rail passengers.


In Chicago, the CTA often operates on two-track ROWs making express service difficult if not impossible.  The CTA two-track operation within the Congress Expressway could be expanded to four tracks to accommodate additional passengers and service levels.  As advocated by Mac Kaye and Mumford, parking could be expanded at the end of the lines to facilitate transfers to and from CTA trains.  While a complete comprehensive BCA has not been done, we know from research that many if not most commuters and other travelers will make the mode change if expanded transit service is fast, frequent and convenient.  The level of time savings during peak congestion periods of such a scenario could be substantive, although these benefits may not be significant or evident during non-peak periods.  However, the benefits to commuters and society as a whole are potentially tremendous in terms of reduced transportation costs, emissions, accidents, and resource consumption.


We also know from research that induced travel demand is substantive with roadway expansion and vice versa.  It may not always be clear just how much these changes will be due in part to differing study results.  Nevertheless, the best alternative for a project can become very clear if pricing reform becomes part of transportation policy.  Charges can be in the form of congestion pricing of expressways, including high-occupancy toll (HOT) lanes, using variable rates consistent with demand.  The downside is that drivers may seek non-expressway alternatives which add to congestion in other areas.  Another option is increased fuel taxes at rates to achieve planning objectives in accordance with known travel demand price elasticities.  An additional alternative, which many consider optimal, is the use of VMT user fees to charge vehicles by distance driven.  If used wisely, apart from implementation costs, pricing programs do not need to be an additional cost in the BCA analysis if transportation subsidies are reduced accordingly from sales/property taxes or other non-user fees.


Rigorous BCA could reveal a quantified monetary advantage to society by downsizing an expressway.  In turn, opportunities are presented for linear open space including bicycle and pedestrian paths to increase use of those modes.  By receiving more accurate pricing signals, the regional population would alter decision-making by moving closer to employment and other amenities.


A difficult question is the level of expressway capacity that should be reduced to optimize economic efficiency.  Generally, a significant number of vehicular trips consist of discretionary trips that do not necessarily need to be taken at all.  In other words, they likely could be combined with other trips or by using other modes based on their importance and the available choices.  But expressway capacity may need to be retained for trips that cannot be eliminated through residential/employment location decision-making such as trucking, buses, and other commercial transportation.  Although, a substantial number of these commercial trips could be diverted to outlying expressways bypassing the densest portions of the urbanized area.  Alternatively, many of them could be diverted to select major arterials, some of which might include expansions and modifications, including intelligent transportation system (ITS) strategies, to facilitate more constant vehicular movement.


1.  Norman Bel Geddes.  Magic Motorways.  1940.  Random House, American Book-Stratford Press, Inc., NY, 209-211.

2.  Robert Gogelson.  Downtown, It's Rise and Fall, 1880-1950.  Yale University Press, 2001.

3.  Lewis Mumford.  The Highway and the City, 1964.  Seeker and Warburg.

4.  Lewis Mumford.  Urban Prospect.  1968.  Hardcourt, Brace and Worle, 98.

5.  Randall Arendt.  Rural By Design, 1994.  American Planning Association, Chicago, 129-132.  Citing sketch by Benton Mac Kaye from Dartmouth College Archives; Benton Mac Kaye.  The Lesson of the Mohawk Trail, 1928.  Zone the State Highways.

6.  Gogelson citing Gilles Duranton and Matthew Turner.  The Fundamental Law of Road Congestion:  Evidence from U.S. Cities.  2011.  American Economic Review, 101(6): 2616-52.

7.  Peter Newman and Jeffrey Kenworthy.  Sustainability and Cities, 1999.  Island Press, Washington, D.C., 111-113.

bottom of page