Gentraf

www.vtpi.
org
Info@vtpi.org
250-508-5150
Generated Traffic and Induced Travel

Implications for Transport Planning
30 June 2023
Todd Litman
Victoria Transport Policy Institute
Abstract
Traffic congestion tends to maintain equilibrium; traffic volumes increase until congestion
delays discourage additional peak-period trips. If road capacity expands, peak-period
trips increase until congestion again limits further traffic growth. The additional travel is
called “generated traffic.” Generated traffic consists of diverted traffic (trips shifted in
time, route and destination), and induced vehicle travel (shifts from other modes, longer
trips and new vehicle trips). Generated traffic often fills a significant portion of capacity
added to congested urban road.
Generated traffic has three implications for transport planning. First, it reduces the
congestion reduction benefits of road capacity expansion. Second, it increases many
external costs. Third, it provides relatively small user benefits because it consists of
vehicle travel that consumers are most willing to forego when their costs increase. It is
important to account for these factors in analysis. This paper defines types of generated
traffic, discusses generated traffic impacts, recommends ways to incorporate generated
traffic into evaluation, and describes alternatives to roadway capacity expansion.
A version of this paper was published in the ITE Journal, Vol. 71, No. 4, Institute of Transportation
Engineers (www.ite.org), April 2001, pp. 38-47.
Todd Litman  1998-2023

You are welcome and encouraged to copy, distribute, share and excerpt this document and its ideas, provided
the author is given attribution. Please send your corrections, comments and suggestions for improvement.
Generated Traffic and Induced Travel: Implications for Transport Planning
Contents
Introduction ..................................................................................................................................... 2
Defining Generated Traffic and Induced Vehicle Travel ................................................................. 3
Measuring Generated Traffic and Induced Vehicle Travel .............................................................. 6
Modeling Generated Traffic and Induced Travel .......................................................................... 11
Land Use Impacts .......................................................................................................................... 13
Costs of Induced Travel ................................................................................................................. 14
Calculating Consumer Benefits...................................................................................................... 17
Emission Impacts ........................................................................................................................... 19
Example ......................................................................................................................................... 20
Counter Arguments ....................................................................................................................... 24
Alternative Transport Improvement Strategies ............................................................................ 26
Legal Issues .................................................................................................................................... 27
Conclusions.................................................................................................................................... 28
References and Information Resources ........................................................................................ 29
1
Introduction
Traffic engineers often treat traffic as a liquid that must flow through the road system, but
urban traffic often behaves more like a gas that expands to fill available space (Jacobsen 1997).
Traffic congestion tends to maintain equilibrium: traffic volumes increase to the point that
congestion delays discourage additional peak-period vehicle trips. Expanding congested roads
attracts latent demand, trips from other routes, times and modes, and encourages longer and
more frequent travel. This is called generated traffic, referring to additional peak-period vehicle
traffic on a particular road. This consists in part of induced vehicle travel, which refers to
absolute increases in vehicle miles travel (VMT) compared with what would otherwise occur
(Hills 1996; Schneider 2018).
Generated traffic reflects the “law of demand,” which states that a good’s consumption
generally increases as its price declines. Roadway improvements that reduce the user costs (i.e.,
the price) of driving encourage more vehicle travel. In the short-run generated traffic represents
a shift along the demand curve; reduced congestion reduces travel time and vehicle operating
costs. Over the long run it represents an outward shift in the demand curve as transport systems
and land use patterns become more automobile dependent, so people must drive more to
maintain a given level of accessibility to goods, services and activities (Deakin, et al. 2020).
This is not to ignore roadway expansion benefits, but generated traffic affects their nature.
Accurate transport planning and project appraisal considers these three effects:
1. Generated traffic reduces the predicted congestion reduction benefits of road capacity expansion (a
type of rebound effect).
2. Induced travel increases many costs, including user expenses, downstream congestion, crashes,
parking costs, pollution, and other environmental impacts. Many of these costs are external and
therefore inefficient and unfair.
3. The additional vehicle traffic provide relatively modest user benefits since it consists of marginal
value vehicle-miles that consumers are most willing to forego if their costs slightly increase.
Ignoring these factors distorts planning decisions (Goodwin and Hopkinson (2023). Experts
conclude, “…the economic value of a scheme can be overestimated by the omission of even a
small amount of induced traffic. We consider this matter of profound importance to the value-
for-money assessment of the road programme” (SACTRA 1994). “…quite small absolute changes
in traffic volumes have a significant impact on the benefit measures…the proportional effect on
scheme Net Present Value will be greater still” (Mackie, 1996), and “The induced travel effects of
changes in land use and trip distribution may be critical to accurate evaluation of transit and
highway alternatives” (Johnston, et al. 2001). Metz (2021) found that, expanding London’s M25
motorway increased traffic volumes up to 23% two to three years after opening, but contrary to
projections, failed to increase traffic speeds, reducing expected economic benefits.
This report describes how generated traffic can be incorporated into transport planning. It
defines different types of generated traffic, discusses their impacts, and describes ways to
incorporate generated traffic into transport modeling and planning, and provides information
on strategies for using existing roadway capacity more efficiently.
2
Defining Generated Traffic and Induced Vehicle Travel

Generated traffic is the additional peak-period vehicle traffic that results from a road
improvement, particularly urban roadway expansions. Congested roads cause people to defer
less-urgent trips, change modes and destinations, and forego avoidable trips. Generated traffic
consists of diverted travel (shifts in time and route) and induced travel (increased total motor
vehicle travel). Highway expansion can stimulate sprawl (dispersed, automobile-dependent
development) which further increasing per capita vehicle travel.
Below are examples of decisions that generate traffic:

 Consumers choose closer destinations when roads are congested and further destinations
when traffic flows more freely. “I want to try the new downtown restaurant but traffic is a
mess now. Let’s just pick up something at the local deli.” This also affects long-term
decisions. “We’re looking for a house within 40-minute commute time of downtown. With
the new highway open, we’ll considering anything as far as Midvalley.”
 Travelers shift modes to avoid driving in congestion. “The post office is only five blocks away
and with congestion so bad this time of day, I may as well walk there.”
 Longer trips may seem cost effective when congestion is light but not when congestion is
heavy. “We’d save $5 on that purchase at the Wal-Mart across town, but it’s not worth
fighting traffic so let’s shop nearby.”
Extensive research indicates that people tend to have fixed travel time budgets, called
Marchetti’s constant (Levinson and Kumar 1997; Litman 2021; Marchetti 1994). Regardless of
conditions people devote about 75 daily minutes to personal (Ahmed and Stopher 2014). As a
result, when travel speeds increase, so do their travel distances. Roadway improvements that
increase traffic speeds tend to induce additional vehicle travel over the long run (Krol 2020). It is
therefore inappropriate to assume that roadway improvements provide travel time savings;
instead their benefits tend to result from the ability to travel to more distant destinations, for
example, to accept a longer distance commute or travel to a more distant holiday destination.
Definitions
Generated Traffic: Additional peak-period vehicle trips on a particular roadway that occur when capacity is
increased. This may consist of shifts in travel time, route, mode, destination and frequency.
Induced travel: An increase in total vehicle mileage due to roadway improvements that increase vehicle trip
frequency and distance, but exclude travel shifted from other times and routes.
Latent demand: Additional trips that would be made if travel conditions improved (less congested, higher
design speeds, lower vehicle costs or tolls).
Triple Convergence: Increased peak-period vehicle traffic volumes that result when roadway capacity
increases, due to shifts from other routes, times and modes.
This is true of roadway expansions intended to reduce traffic congestion. Traffic congestion
tends to maintain equilibrium: it increases to the point that delays discourage additional peak-
period trips. If congested roads are expanded, motorists will make additional peak-period trips
that they would otherwise forego, driving additional vehicle-miles.
3
Figure 1 illustrates this pattern. Traffic volumes grow until congestion develops, then the growth
rate declines and achieves equilibrium, indicated by the curve becoming horizontal. A demand
projection made during this growth period will indicate that more capacity is needed, ignoring
the tendency of traffic volumes to eventually level off. If additional lanes are added there will be
another period of traffic growth as predicted.
Figure 1 How Road Capacity Expansion Generates Traffic
Traffic Volume With Added Capacity Traffic grows when roads are
Traffic Volume Without Added Capacity uncongested, but the growth
rate declines as congestion
Traffic Lanes and Volume
2
develops, reaching a self-limiting
equilibrium (indicated by the
Projected curve becoming horizontal). If
Traffic
Generated capacity increases, traffic grows
Growth
Traffic until it reaches a new
1 equilibrium. This additional
peak-period vehicle travel is
called “generated traffic.” The
portion that consists of absolute
increases in vehicle travel (as
opposed to shifts in time and
0 route) is called “induced travel.”
Roadway
Time ----> Capacity
Added
Generated traffic can be considered from two perspectives. Highway planners are primarily
concerned with the traffic generated on the expanded road segment, since this affects the
project’s congestion reduction benefits. A broader perspective is concerned with changes in
total vehicle travel (induced travel) that affect overall benefits and costs. Table 1 describes
various types of generated traffic. In the short term, most generated traffic consists of trips
diverted from other routes, times and modes, called Triple Convergence (Downs 1992). Over the
long term an increasing portion is induced travel. In some situations, adding roadway capacity
can reduce overall network efficiency, called Braess’s Paradox (Youn, Jeong and Gastner 2008).
Highway capacity expansion can induce additional vehicle travel on adjacent roads by
stimulating more dispersed, automobile-dependent development (Hansen, et al. 1993).
Although these indirect impacts are difficult to quantify they are potentially large and should be
considered in transport policy and planning analysis (Byun, Park and Jang 2017).
4
Table 1 Types of Generated Traffic

Time Travel Cost
Type of Generated Traffic Category Frame Impacts Impacts
Shorter Route – Improved road allows drivers to use Small
more direct route. Diverted trip Short term reduction Reduction
Longer Route – Improved road attracts traffic from
more direct routes. Diverted trip Short term Small increase Slight increase
Time Change – Reduced peak period congestion
reduces the need to defer trips to off-peak periods. Diverted trip. Short term None Slight increase
Mode Shift; Existing Travel Choices – Improved traffic
flow makes driving relatively more attractive than other Induced Increased Moderate to
modes. vehicle trip Short term driving large increase
Mode Shift; Changes in Travel Choice – Less demand Increased
leads to reduced rail and bus service, less suitable driving,
conditions for walking and cycling, and more automobile Induced reduced Large increase,
ownership. vehicle trip Long term alternatives reduced equity
Destination Change; Existing Land Use – Reduced travel
costs allow drivers to choose farther destinations. No Moderate to
change in land use patterns. Longer trip Short term Increase large increase
Destination Change; Land Use Changes – Improved More driving Moderate to
access allows land use changes, especially urban fringe and auto large increase,
development. Longer trip Long term dependency equity costs
New Trip; No Land Use Changes – Improved travel time
allows driving to substitute for non-travel activities. Induced trip Short term Increase Large increase
Automobile Dependency – Synergetic effects of Increased
increased automobile oriented land use and driving, fewer Large increase,
transportation system. Induced trip Long term alternatives reduced equity
Some types of generated traffic represent diverted trips (trips shifted from other times or routes)
while others increase total vehicle travel, reduce travel choices, and affect land use patterns.
What constitutes short- and long-term impacts can vary. Some short term effects, such as mode
shifts, may accumulate over several years, and some long term effects, such as changes in
development patterns, can begin almost immediately after a project is announced if market
conditions are suitable. Roadway expansion impacts tend to include:
 First order. Reduced congestion delay, increased traffic speeds.
 Second order. Changes in time, route, destination and mode.
 Third order. Land use changes. More dispersed, automobile-oriented development.
 Fourth order. Overall increase in automobile dependency. Degraded walking and cycling
conditions (due to wider roads and increased traffic volumes), reduced public transit service
(due to reduced demand and associated scale economies, sometimes called the Downs-Thomson
paradox), and social stigma associated with alternative modes (Noland and Hanson 2013, p. 75).
Such impacts can also occur in reverse: reducing urban roadway capacity often reduces total
vehicle travel (Cairns, Hass-Klau and Goodwin 1998; CNU 2011; ITDP 2012; ITF 2021) which is
sometimes called traffic evaporation (EC 2004).
5
Measuring Generated Traffic and Induced Vehicle Travel

Numerous studies using various analysis methods have quantified generated traffic and induced
travel impacts (Deakin, et al. 2020; WSP 2018). Their findings are summarized below:
 The National Center for Sustainable Transportation’s Induced Travel Calculator (NCST 2019)
estimates the incremental vehicle travel induced by adding general-purpose or high-occupancy-
vehicle (HOV) lane miles to roadways. It is calibrated for California’s urbanized counties, but the
methodology is transferable to other geographic areas.
 Sophisticated analyses of 545 European cities indicates that urban highway expansion tends to
increase vehicle traffic and so fails to solve congestion (Garcia-López, Pasidis, and Viladecans-
Marsal 2020). The study indicates that each 1% increase in highway lane-kilometers typically
increases total vehicle kilometers by 1.2%. The analysis found significantly less congestion
(indicated by vehicle-kms relative to the log of lane-kms) in cities with road pricing and high
quality rail transit. A 1% increase in lane kilometers increases congestion by 1.9% in cities
without highway tolls but only 0.3% in cities with tolls. A 1% increase in railroad network length
decreases congestion by 0.6% in a city without subways, 0.8% in a city with the average share of
subways, and 1.3% in a city where the majority of the railroad network consists of subways.
 The report, The Congestion Con: How More Lanes and More Money Equals More Traffic (TfA
2020) analyzed how roadway expansions affected per capita congestion delay in the 100 largest
urbanized areas in the U.S. between 1993 and 2017. During that period governments spent
more than $500 billion on highway projects but congestion grew 144%, far more than
population, and the regions that expanded roads the most tended to have more congestion
growth than those that expanded less. The authors concluded that this resulted from generated
traffic which filled the added capacity. plus the long-term effects of increased sprawl and
increased per capita vehicle travel induced by the additional roadway capacity.
 Detailed analysis by Hymel (2019) found that U.S. vehicle miles traveled increase in proportion
with lane-mileage, and capacity expansion congestion relief generally vanishes within five years.
 A Statistical Model of Regional Traffic Congestion in the United States (Marshall 2016) used real-
time traffic data to analyze factors that affected congestion in 74 U.S. urban regions. It found
that more arterial capacity is related to less congestion but more freeway capacity is not. It
found that congestion increases with incomes indicating that economic productivity attracts
population growth, which also increases congestion. The study concludes that in congested
urban areas, arterial expansion may reduce congestion but freeway expansions do not.
 Graham, McCoy and Stephens (2014) quantify roadway capacity expansion effects on aggregate
urban traffic volume and density in U.S. cities using a mixed model propensity score estimator
which accounts for confounding unobserved characteristics. They found that a 10% increase in
lane miles increases average VMT 9% beyond ‘natural growth.’ They conclude that even major
urban highway expansions can provide little or no long-term congestion reductions.
 A review by Handy and Boarnet (2014) found that short-run highway expansion elasticities
generally range from 0.3 to 0.6, and long-run effects typically range from 0.6 to just over 1.0,
meaning that each 10% increase in road capacity increases traffic volumes by 3-6% within two
years, and 6-10% after about five years. They found that more recent studies using more
sophisticated methodologies tend to find higher elasticities. They conclude that expanding
congested urban highways is unlikely to reduce long run congestion or associated GHGs.
6
 Duranton and Turner (2008) found that in U.S. urban regions, vehicle travel increases
proportionately to highway capacity due to four effects: increased driving by current residents,
an inflow of new residents, and more transport intensive production activity. They conclude
that, without congestion pricing, increasing road or public transit supply is unlikely to relieve
congestion, and current roadway supply exceeds optimums.
 Cervero (2003a & 2003b) used data on freeway capacity, traffic volumes, demographic and
geographic factors in California between 1980 and 1994. He estimated a 0.64 long-term elasticity
of VMT with respect to traffic speed, meaning that a 10% speed increases VMT 6.4%, about a
quarter of which results from land use changes (e.g., additional urban fringe development). He
estimated that about 80% of additional roadway capacity is filled with additional peak-period
travel, about half of which (39%) can be considered the direct result of the added capacity.
 Noland (2001) and (Noland and Lem 2002) used time-series travel data for various roadway
types to evaluate induced travel. They found an elasticity of vehicle travel with respect to lane
miles of 0.5 in the short run, and 0.8 in the long run. This means that half of increased roadway
capacity is filled with added travel within about 5 years, and that 80% of the increased roadway
capacity will be filled eventually.
 Leading U.K. transportation economists concludes that the elasticity of travel volume with
respect to travel time is -0.5 in the short term and -1.0 over the long term (SACTRA 1994). This
means that reducing travel time on a roadway by 20% typically increases traffic volumes by 10%
in the short term and 20% over the long term. The following are elasticity values for vehicle
travel with respect to travel time: urban roads, short-term -0.27, long term –0.57; rural roads,
short term –0.67, long term –1.33 (Goodwin 1996).
 Noland and Quddus (2006) found that increases in road space or traffic signal control systems that
smooth traffic flow tend to induce additional vehicle traffic which quickly diminish any initial
emission reduction benefits.
 Tennøy, Tønnesen and Gundersen (2019) found that Norwegian highway expansions provide
only short-term congestion relief, and by increasing sprawled development, increase total traffic
growth. They find that road authorities generally overlooked these effects.
 Cervero and Hanson (2000) found the elasticity of VMT with respect to lane-miles to be 0.56,
and an elasticity of lane-miles with respect to VMT of 0.33, indicating that roadway capacity
expansion results in part from anticipated traffic growth.
 A comprehensive study found that in the U.S., a 10% increase in urban road density (lane-miles
per square mile) increases per capita annual VMT by 0.7% (Barr 2000).
 Yao and Morikawa (2005) analyzed the travel induced by high speed rail improvements between
major Japanese cities. They calculate elasticities of induced travel (trips and VMT) with respect
to fares, travel time, access time and service frequency for business and nonbusiness travel.
 Odgers (2009) found that Melbourne, Australia freeway traffic speeds did not increase as
predicted following highway construction, apparently due to induced traffic. He concludes that,
“major road infrastructure initiatives and the consequent economic investments have not yet
delivered a net economic benefit to either Melbourne’s motorists or the Victorian community.”
7
 Burt and Hoover (2006) found that each 1% increase in road lane-kilometres per driving-age
person increases per capita light truck travel 0.49% and car travel 0.27%, although they report
that these relationships are not statistically significant, falling just outside the 80% confidence
interval for cars and the 90% confidence interval for light trucks.
 Hymel, Small and Van Dender (2010) used 1966-2004 U.S. state-level cross-sectional time series
data to evaluate how income, fuel price, road supply and traffic congestion affect vehicle miles
travel (VMT). They find the elasticity of VMT with respect to statewide road density is 0.019 in
the short run and 0.093 in the long run (a 10% increase in total lane-miles per square mile
increases state vehicle mileage by 0.19% in the short run and 0.93% in the long run); with
respect to total road miles is 0.037 in the short run and 0.186 in the long run (a 10% increase in
lane-miles causes state VMT to increase 0.37% in the short run and 1.86% over the long run);
and vehicle use with respect to congestion is -0.045 (a 10% increase in total regional congestion
reduces regional VMT 0.45% over the long run), but this increases with income, assumedly
because the opportunity cost of time increases with wealth. Their analysis indicates that long-
run travel elasticities are typically 3.4–9.4 times short-run elasticities.
 Using sophisticated statistical analysis of traffic flow in 24 cities, Anupriya, Bansal and Graham
(2023) found that increasing road network capacity does not substantially increase average travel
speeds, and by increasing total traffic volumes can increase overall traffic disbenefits.
 The Handbook of Transportation Engineering finds that urban highway capacity expansion often
fails to significantly increase travel speeds due to latent demand (Kockelman 2010). They conclude
that the long-run elasticities of VMT with respect to roadspace is generally 0.5 to 1.0 after
controlling for population growth and income, with values of almost 1.0, suggesting that new
roadspace is totally filled by generated traffic where congestion is relatively severe.
 A meta-analysis by Schiffer, Steinvorth and Milam (2005) reached the following conclusions:
o Induced travel effects exist – The elasticity of VMT with respect to added lane-miles or
reductions in travel time is generally greater than zero and the effects increase over time.
Figure 3 summarizes their results.
o Short-term induced travel effects are smaller than long-term effects – As measured by the
increase in VMT with respect to an increase in lane-miles, short-term effects have an
elasticity range from near zero to about 0.40, while long-term elasticities range from about
0.50 to 1.00. This means that a 10% increase in lane-miles can cause up to a 4% increase in
VMT in the short term and a 10% increase in the long term.
o Induced travel effects generally decrease with the size of the unit of study – Larger effects are
measured for single facilities while smaller effects are measured for regions and subareas.
This is mainly due to diverted trips (drivers changing routes) causing more of the change on a
single facility, whereas, at the regional level, diverted trips between routes within the region
are not considered induced travel unless the trips become longer as a result.
o Traditional four-step travel demand models do not fully address induced travel or induced
growth – Land use allocation methods overlook accessibility effects, trip generation often
fails to account for latent trips (potential trips constrained by congestion), many models
overlook time-of-day shifts, and static traffic assignment algorithms may not account for
queuing impacts on route shifts; all of which underestimate generated traffic effects.
8
Figure 3 VMT With Respect to Road Capacity (Schiffer, Steinvorth and Milam 2005)
This figure
summarizes long term
vehicle travel
elasticities with
respect to roadway
capacity.
 Melo, Graham and Canavan (2012) found a positive relationship between urban highway
expansion and vehicle travel in the U.S. between 1982 and 2009.
 Rahmana, Bakerb and Rahmanc (2020) found that in Dhaka, Bangladesh, urban intersection
flyovers typically provide a one-minute time savings, which increased affected vehicle trip
generation by 35%.
 Özuysal and Tanyel (2008) found the elasticity of travel per vehicle relative to Turkish state
highway supply is 2.0 for private vehicles and 3.5 for commercial vehicles over 3-5 year periods.
 Analysis by Professor Ismail Sahin of Turkey’s Yildiz Technical University shows that after new
bridges were built in Istanbul, traffic volumes increased, representing induced vehicle trips,
resulting in a new, higher level of congestion equilibrium.
Figure 4 Istanbul Bridge Traffic Volumes (Personal correspondence with

Professor Ismail Sahin 2015)
This graph shows Average

Annual Daily Trips (AADTs)
on Bosporus Bridge (blue
bars) opened in 1973 and
Fatih Sultan Mehmet (FSM)
Bridge (orange bars) opened
in 1988. This shows the
tendency of traffic volumes
to fill the new capacity and
then reach equilibrium as
congestion becomes self-
limiting.
9
The amount of traffic generated by a road project varies depending on conditions. It is not
capacity expansion itself that generates travel, it is the reduction in delay and therefore per-mile
travel costs (Milam, et al. 2017). Expanding uncongested roads generates no traffic, although
paving a dirt road or significantly raising roadway design speeds may induce more vehicle travel.
In general, the more congested a road, the more traffic is generated by expansions. Increased
capacity on highly congested roads often generates considerable traffic. Older studies of the
elasticity of VMT growth with respect to increased roadway lane-miles performed during the
early years of highway building (during the 1950s through 1970s) have little relevance for
evaluating current urban highway capacity expansion. In developed countries, where most
highway expansion now occurs on congested links, such projects are likely to generate
considerable amounts of traffic, providing only temporary congestion reduction benefits.
Gridlock?
People sometimes warn of roadway gridlock without some recommended action, such as roadway expansion. Such
claims are usually exaggerated because they ignore traffic congestion’s tendency toward equilibrium. Gridlock is a
specific condition that occurs when backups in a street network block intersections, stopping traffic flow. Gridlock
can be avoided with proper intersection design and traffic law enforcement. Increasing regional highway capacity
can increase this risk by adding more traffic to surface streets where gridlock occurs.
Generated traffic usually accumulates over several years. Under typical urban conditions, more
than half of added capacity is filled within five years of project completion by additional vehicle
trips that would not otherwise occur, with continued but slower growth in later years (Goodwin
1998). Figure 5 shows typical generated traffic growth based on various studies. Techniques for
modeling these impacts are described in the next section.
Figure 5 Elasticity of Traffic Volume With Respect to Road Capacity
This illustrates traffic growth on a road after its capacity increases. About half of added capacity
is typically filled with new traffic within a decade of construction. (Based on cited studies)
10
Modeling Generated Traffic and Induced Travel

Traffic congestion tends to maintain equilibrium: it increases to the point that delays cause
some travellers to shift time, mode, route or destination. To predict generated traffic transport
models must incorporate “feedback” reflecting how congestion affects travel behavior including
long-term changes in transport and land use patterns. Because these relationships are non-
linear, small amounts of generated traffic can significantly affect roadway project cost efficiency.
All current traffic models can predict route and mode shifts, and some can predict changes in
trip frequency, scheduling and destination, but few incorporate feedback on long-term effects
such as the tendency of highway expansions to increase automobile-dependent urban fringe
development where households own more vehicles and drive more annual miles than they
would if located in more central, multimodal areas (Milam, et al. 2017; Næss, Nicolaisen and
Strand 2012). As a result, current models overestimate highway expansion costs and
underestimate long-term induced vehicle travel, and associated costs, including downstream
traffic and parking congestion, crashes and pollution emissions (Deakin, et al. 2020).
Volker, Lee, and Handy (2020) examined the evaluation methods used in five recent highway
project. They found that conventional analyses frequently fail to account for induced travel
effects, which exaggerated their benefits and underestimated their environmental costs. The
authors used this information to develop the National Center for Sustainable Transportation’s
Induced Travel Calculator, which estimates the incremental vehicle travel induced by adding
general-purpose or high-occupancy-vehicle (HOV) lane miles to roadways. The Rocky Mountain
Institute used this methodology to develop the SHIFT Calculator which predicts induced vehicle
travel and emissions from capacity expansions of large roadways for U.S. urban regions.
Ramsey (2005) found that a suburban highway expansion project’s net benefits declined 50% if
the project caused just 2% of the regional population to move from urban to suburban
locations. Similarly, Næss, Nicolaisen and Strand (2012) found that ignoring some induced traffic
effects significantly affected the estimated value of a proposed Copenhagen, Denmark highway
expansion. When induced travel was considered the results show lower travel time savings and
more adverse environmental impacts, resulting in significantly lower benefit-cost ratio. They
conclude that, “By exaggerating the economic benefits of road capacity increase and
underestimating its negative effects, omission of induced traffic can result in over-allocation of
public money on road construction and correspondingly less focus on other ways of dealing with
congestion and environmental problems in urban areas.” Analysis of urban highway expansion
impacts on total emissions by Williams-Derry (2007) indicates that construction and induced
vehicle travel emission quickly exceed any emission reductions from less congestion.
A study, Climate Emissions Analysis: Metro’s Indirect Impact on Greenhouse Gas Emissions
(METRO 2022) found that the emissions reduced by public transit and active transportation
improvements in Los Angeles are more than offset by planned freeway expansions, particularly
over the long run.
Transportation modelers have developed techniques for incorporating full feedback (SACTRA
1994; Loudon, Parameswaran and Gardner 1997; Schiffer, Steinvorth and Milam 2005). This
recognizes that expanding the capacity of congested roads increases the number and length of
trips in a corridor (DeCorla-Souza and Cohen 1999). Federal clean air rules require that these
techniques be used in metropolitan transportation models to evaluate the effects transport
11
system changes have on vehicle emissions, but many metropolitan planning organizations have
yet to comply, and few models used in medium and small cities have full feedback. Full feedback
is necessary to accurately predict future congestion and traffic speeds, and the incremental
costs and benefits of alternatives. Models that lack feedback tend to overestimate future
congestion problems and overestimate capacity expansion benefits.
Models that fail to consider generated traffic were found to overvalue roadway capacity
expansion benefits by 50% or more (Williams and Yamashita 1992). Another study found that
the ranking of preferred projects changed significantly when feedback is incorporated into
project assessment (Johnston and Ceerla 1996). Ignoring generated traffic tends to skew
planning decisions toward highway projects and away from No Build and transportation demand
management alternatives such as road pricing, transit improvements and commute trip
reduction programs. UK Department for Transport’s Transport Analysis Guidance (DfT 2007),
includes a section on Variable Demand Modelling
(www.dft.gov.uk/webtag/documents/expert/unit3.10.1.php) which describes methods for
incorporating induced travel demand into project appraisal.
Short Cut Methods of Incorporating Induced Demand

Based on comments by Phil Goodwin, 2001.
The easiest way to incorporate induced demand into conventional traffic models is to apply an overall
demand elasticity to forecasted changes in travel speed, calculated either:
 Elasticities applied to generalized costs (travel time and financial costs) using a price elasticity (about -
0.3 for equilibrium, less for short term), with monetized travel time costs. The time elasticity is
generally about -0.5 to -0.8 or so, though this is highly dependent on context. Where to apply it
depends on the model used. With a fixed trip matrix altered only by reassignment, apply elasticities to
each separate cell, or the row and column totals, or the overall control total - depending on how short
the short cut has to be. Or add a separate test at the end.
or
 Direct application of a ‘capacity elasticity,’ i.e. percent change in vehicle miles resulting from a 1%
change in highway capacity, for which lane miles is sometimes used as a proxy, the elasticity in that
case usually coming out at about -0.1. This will tend to underestimate the effect if the capacity
increase is concentrating on bottlenecks.
Care is needed if the basic model has cost-sensitive distribution and mode split, as this will already
account for some induced traffic. Induced traffic consists of several types of travel changes that make
vehicle miles “with” a scheme different from “without,” including re-assignment to longer routes and
increased trip generation. Although time-shifting is not induced traffic, it has similar effects on congestion
reduction benefits and is often a large response. Ideally you iterate on speed and allow for the effect from
retiming of journeys, and separate the various behavioural responses which make up induced traffic.
These short cuts are subject to bias, but less than the bias introduced by assuming zero induced traffic.
12
Land Use Impacts

An important issue related to generated and induced travel is the degree to which roadway
improvements affect land use patterns, and in particular, whether highway capacity expansion
stimulates lower-density, urban fringe development (i.e., urban sprawl), and the costs to society
that result (Deakin, et al. 2020; USEPA 2001; ICF Consulting 2005). Land use changes are one
category of induced travel. Such changes take a relatively long time to occur, and are influenced
by additional factors, but they are durable effects with a variety of economic, social and
environmental impacts.
Urban economists have long realized that transportation can have a major impact on land use
development patterns, and in many situations improved accessibility can stimulate development
location and type. Different types of transportation improvements tend to cause different types
of land use development patters: highway improvements tend to encourage lower-density,
automobile-oriented development at the urban fringe, while transit improvements tend to
encourage higher-density, multi-modal, urban redevelopment, although the exact types of
impacts vary depending on specific conditions and the type of transportation improvements
implemented (Rodier, Abraham, Johnston and Hunt 2001; Boarnet and Chalermpong 2002).
Some researchers claim that investing in road construction does not lead to the sprawl (Hartgen
2003), although the evidence indicates otherwise. Even in relatively slow-growth regions with
modest congestion problems, highway expansions increase suburban development by 15-25%.
These effects are likely to be much greater in large cities with significant congestion, where
peak-period traffic congestion limits commute trip distances, and increased roadway capacity
would significantly improve automobile access to urban fringe locations. This is particularly true
if the alternative is to implement Smart Growth development policies and improved walking,
cycling and transit transportation.
There has been considerable debate over the benefits and costs of sprawl and Smart Growth.
Table 2 summarizes some benefits that tend to result from reduced sprawl.
Table 2 Smart Growth Benefits (Ewing and Hamidi 2014; Litman 2016)
Economic Social Environmental
Reduced development and public Greenspace and wildlife habitat
service costs. preservation.
Consumer transportation cost Improved transportation choice, Reduced air pollution.
savings. particularly for nondrivers. Reduce resource consumption.
Economies of agglomeration. Improved housing choices. Reduced water pollution.
More efficient transportation. Community cohesion. Reduced “heat island” effect.
13
Costs of Induced Travel

Driving imposes a variety of costs, including many that are external, that is, not borne directly by
users. Table 3 illustrates one estimate of the magnitude of these costs. Other studies show
similar costs, with average values of 10-30¢ per vehicle-kilometer, and more under urban-peak
conditions (Litman 2003).
Table 3 Motor Vehicle Indirect and External Costs (Delucchi 1996)

Cost Item Examples Vehicle-Year Vehicle-Mile
Bundled private sector costs Parking funded by businesses $337-1,181 2.7-9.4 cents
Public infrastructure and Public roads, parking funded by
services local governments $662-1,099 5.3-8.8 cents
External crash damages to vehicles,
Monetary externalities medical expenses, congestion. $423-780 3.4-6.2 cents
Environmental damages, crash
Nonmonetary externalities pain. $1,305-3,145 10.4-25.2 cents
Totals $2,727-6,205 22-50 cents
This table summarizes an estimate of motor vehicle indirect and external costs. (US 1991 Dollars)
Any incremental external costs of generated traffic should be included in project evaluations,
“incremental” meaning the difference between the external costs of the generated travel and
the external costs of alternative activities. For diverted traffic this is the difference in external
costs between the two trips. For induced travel this is the difference in external costs between
the trip and any non-travel activity it replaces, which tends to be large since driving has greater
external costs than most other common activities. Most generated traffic occurs under urban-
peak travel conditions, when motor vehicle external costs are greatest, so incremental external
costs tend to be high.
Incremental external costs depend on road system conditions and the type of generated traffic.
Generated traffic often increases downstream congestion (for example, increasing capacity on a
highway can add congestion on surface streets, particularly near on- and off-ramps). In some
conditions adding capacity actually increases congestion by concentrating traffic on a few links
in the network and by reducing travel alternatives, such as public transit (Arnott and Small
1994). Air emission and accident rates per vehicle-mile may decline if traffic flows more freely,
but these benefits decline over time and are usually offset as generated traffic leads to renewed
congestion and increased vehicle travel (TRB 1995; Cassady, Dutzik and Figdor 2004).
Table 4 compares how different types of generated traffic affect costs. All types reduce user
travel time and vehicle costs. Diverted trips have minimal incremental costs. Longer trips have
moderate incremental costs. Shifts from public transit to driving may also have moderate
incremental costs, since transit service has significant externalities but also experiences
economies of scale and positive land use impacts that are lost if demand declines. Induced trips
have the largest incremental costs, since they increase virtually all external costs. Longer and
induced vehicle trips can lead to more automobile dependent transportation and land use over
the long term. These costs are difficult to quantify but are probably significant (Ewing and
Hamidi 2014).
14
Table 4 Cost Impacts of Roadway Capacity Expansion

Costs Reduced Costs Increased
Diverted Trips Longer Trips Induced Trips
 Downstream congestion
 Travel Time
 Road facilities
 Vehicle Operating
Costs  Parking facilities
 Per-mile crash rates (if  Traffic services

 Downstream
implemented in congestion  Per-capita crash rates
conjunction with
roadway design  Road facilities  Pollution emissions
improvements, but  Traffic services  Noise
these are often offset
if traffic speeds  Per-capita crash rates  Resource externalities
increase).  Pollution emissions  Land use impacts
 Per-mile pollution  Noise  Barrier effect
emissions (if
congestion declines,  Resource externalities  Transit efficiency
but these may be
 Land use impacts  Equity
offset if traffic speeds  Downstream
increase). congestion  Barrier effect  Vehicle ownership costs
Increased roadway capacity tends to reduce two costs, but increases others.
The incremental external costs of road capacity expansion tend to increase over time as the
total amount of generated traffic grows and an increasing portion consists of induced motor
vehicle travel and trips.
Table 5 proposes default estimates of the incremental external costs of different types of
generated traffic. These values can be adjusted to reflect specific conditions and analysis needs.
Table 5 Estimated Incremental External Costs of Generated Traffic

Type Description Cost Per Mile
Trips shifted from off-peak to peak, or from
Time and route shift another route. 5 cents
Transit-to-Auto mode shift, Trips shifted from transit to driving alone, and
and longer trips increased automobile trip lengths. 15 cents
Additional motor vehicle trip, including travel
Induced vehicle trip shifted from walking, cycling and ridesharing. 30 cents.
This table indicates the estimated incremental costs of different types of generated traffic.
15
There is considerable debate concerning the emission impacts of roadway expansion (TRB
1995). Although expanding congested roadways may sometimes reduce per-kilometer emission
rates, it generally increases total emissions, particularly over the long run by increasing high
traffic speeds (more than 80 kms/hr), and by inducing more vehicle travel. According to a study
by the Norwegian Centre for Transport Research (TØI 2009):
“Road construction, largely speaking, increases greenhouse gas emissions, mainly

because an improved quality of the road network will increase the speed level, not the
least in the interval where the marginal effect of speed on emissions is large (above
80km/hr). Emissions also rise due to increased volumes of traffic (each person traveling
further and more often) and because the modal split changes in favor of the private car,
at the expense of public transport and bicycling.”
Table 6 summarizes roadway improvement emission impacts, including effects on emission

rates per vehicle mile, increases in total vehicle mileage, and emissions from road construction
and maintenance activities.
Table 6 Roadway Expansion Greenhouse Gas Emission Impacts (TØI 2009)

General Estimates Large Cities Small Cities Intercity Travel
Emission reductions Short term Depends on Depends on
per vehicle-kilometer reductions. Stable situation, ranging situation. Emissions
due to improved and or some increase from no change may decline or
expanded roads. over the long-term. to large increases. increase.
Increased vehicle
mileage (induced A 10% reduction in
vehicle travel), short travel time increases Significant emission Moderate Moderate emission
term (under five years) traffic 3-5% growth emission growth growth
Increased vehicle
mileage (induced A 10% reduction in
travel), long term travel time increases Significant emission Moderate Moderate emission
(more than five years) traffic 5-10% growth emission growth growth
12 tonnes of CO2
equivalent for 2-lane
Road construction and roads and 21 tonnes Road construction emissions are relatively modest compared
improvement activity for 4-lane roads. with traffic emissions.
33 tonnes of CO2
Roadway operation equivalent for 2-lane
and maintenance roads and 52 tonnes Road operation and maintenance emissions are relatively
activity for 4-lane roads. modest compared with traffic emissions.
This table summarizes roadway improvement emission impacts according to research by the
Norwegian Centre for Transport Research.
16
Calculating Consumer Benefits

Generated traffic represents increased mobility, which provides consumer benefits. However,
these benefits tend to be modest because generated traffic consists of marginal value trips, the
trips that people are most willing to forego. To calculate these benefits economists use the Rule
of Half, which states that the benefits of additional travel are worth half the per-trip saving to
existing travelers, as illustrated in Figure 6 by the fact that B is a triangle rather than a rectangle.
Figure 6 Vehicle Travel Demand Curve Illustrating the Rule-of-Half
Demand Curve
Reduced User Cost

User Travel Costs
A B
Vehicle Travel
Reduced user costs (downward shift on Y axis) increases vehicle travel (rightward shift on X axis).
Rectangle A shows savings to existing trips. Triangle B shows generated travel benefits.
Explanation of the “Rule of Half”

When consumers change their travel in response to a financial incentive, the net consumer surplus averages
half of their price change (called the “rule of half”). Let me illustrate.
Let’s say that by purchasing a hybrid or electric car, your vehicle operating costs decline from 20₵ to 10₵ per
mile, in response you increase 10,000 to 11,000 annual vehicle-miles. The added vehicle-miles have small
incremental value to you, between 0¢ and 10¢. If you consider the additional mile worth less than 0¢ (i.e., it
has no value), you would not take it. If you considered it worth more than 10¢ per mile, you would have driven
that mile without the price reduction. Of the 1,000 miles added we can assume that the average net benefit to
users (called the consumer surplus) is the mid-point of this range, that is, 5¢ per vehicle mile. Thus, we can
calculate the value of the added miles as 5¢ times 1,000 added miles. Conversely, a 10₵ per mile price increase
that reduces vehicle travel by 1,000 miles imposes a net cost to consumers of $50.
Some people complicate this analysis by trying to track individual changes in consumer travel time,
convenience and vehicle operating costs, but that is unnecessary information. All we need to know to value
the net consumer surplus is the perceived change in price, either positive or negative, and the resulting
change in consumption. This incorporates all of the complex trade-offs that consumers make between money,
time, convenience and the value off mobility.
Because induced travel provides relatively small user benefits, and imposes external costs such
as downstream congestion, parking costs, accident risk imposed on other road users, pollution
17
emissions, sprawl and other environmental costs, the ratio of benefits to costs, and therefore
total net benefits of travel, tend to decline as more travel is induced.
Failing to account for the full impacts of generated and induced travel tends to exaggerate the
benefits of highway capacity expansion and undervalue alternatives such as transit
improvements and pricing reforms (Romilly 2004). Some newer project evaluation models, such
as the FHWA’s SMITE and STEAM sketch plan programs, incorporate generated traffic effects
including the Rule of Half and some externalities (FHWA 1998; Wang, Zhong and Hunt 2019).
The benefits of increased mobility are often capitalized into land values. For example, a highway
improvement can increase urban periphery real estate prices, or a highway offramp can increase
nearby commercial land values. Because this increase in land values is an economic transfer
(land sellers gain at the expense of land buyers), it is inappropriate to add increased real estate
values and transport benefits, such as travel time savings (which represent true resource
savings). This would double count benefits.
18
Emission Impacts
Highway expansion advocates sometimes claim that by reducing traffic congestion, such
projects will reduce air pollution emissions, but research indicates that this is generally untrue
(Noland and Quddus 2006). Per-mile emission rates are generally minimized at 20-50 miles per
hour, as indicated in Figures 7 and 8. As a result, reducing extreme congestion (LOS E or F), so
traffic speeds rise above 30 mph may reduce emission rates, but reducing mild congestion (LOS
C or D), so traffic speeds increase above 50 mph are likely to increase emission rates, and if
roadway expansions induce additional vehicle travel they are likely to increase total emissions.
Figure 6 Speed-Emission Curves (Fontaras, et al. 2014)
Kilometers per Hour
Figure 7 U.S. Speed-Emission Curves (Barth and Boriboonsomsin 2009)

Vehicle fuel consumption
and emission rates are
generally minimized at 20-
50 miles per hour. As a
result, reducing extreme
congestion (LOS E or F), so
traffic speeds rise above
30 mph may reduce
emission rates, but
reducing mild congestion
(LOS C or D), so traffic
speeds increase above 50
mph are likely to increase
emission rates
As a result, roadway expansions that reduce extreme congestion may reduce emission rates in
the short run, but these impacts are generally small and more than offset over the long run by
more high-speed driving and induced vehicle travel. In contrast, other congestion reduction
strategies, such as high quality public transit, High Occupancy Vehicle (HOV) lanes, efficient road
pricing, and commute trip reduction programs, provide much greater emission reductions
(Litman 2019).
19
Example
A four-lane, 10-kilometer highway connects a city with nearby suburbs. The highway is
congested 1,000 hours per year in each direction. Regional travel demand is predicated to grow
at 2% per year. A proposal is made to expand the highway to six lanes, costing $25 million in
capital expenses and adding $1 million in annual highway operating expenses.
Figure 9 illustrates predicted traffic volumes. Without the project peak-hour traffic is limited to
4,000 vehicles in each direction, the maximum capacity of the two-lane highway. If generated
traffic is ignored the model predicts that traffic volumes will grow at a steady 2% per year if the
project is implemented. If generated traffic is considered the model predicts faster growth,
including the basic 2% growth plus additional growth due to generated traffic, until volumes
levels off at 6,000 vehicles per hour, the maximum capacity of three lanes.
Figure 9 Projected Traffic

6,000
5,000
Trips Per Peak Hour
4,000
3,000 3 Lanes, Considering Generated Traffic
2,000 3 Lanes, Ignoring Generated Traffic
2 Lanes
1,000
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Years ==>
If generated traffic is ignored the model predicts that traffic volumes will grow at a steady 2%
per year if the project is implemented. If generated traffic is considered the model predicts a
higher initial growth rate, which eventually declines when the road once again reaches capacity
and becomes congested. (Based on the “Moderate Latent Demand” curve from Figure 3)
The model divides generated traffic into diverted trips (changes in trip time, route and mode)
and induced travel (increased trips and trip length), using the assumption that the first year’s
generated traffic represents diverted trips and later generated traffic represents induced travel.
This simplification appears reasonable since diverted trips tend to occur in the short-term, while
induced travel is associated with longer-term changes in consumer behavior and land use
patterns.
Roadway volume to capacity ratios are used to calculate peak-period traffic speeds, which are
then used to calculate travel time and vehicle operating cost savings. Congestion reduction
benefits are predicted to be significantly greater if generated traffic is ignored, as illustrated in
Figure 10.
20
Figure 10 Projected Average Traffic Speeds

100
Av erage Traffic Speed (km/hr)

80
60
40
Ignoring Generated Traffic
20 Considering Generated Traffic
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Years ==>
Ignoring generated traffic exaggerates future traffic speeds and congestion reduction benefits.
Incremental external costs are assumed to average 10¢ per vehicle-km for diverted trips (shifts
in time, route and mode) and 30¢ per vehicle-km for induced travel (longer and increased trips).
User benefits of generated traffic are calculated using the Rule-of-Half.
Three cases where considered for sensitivity analysis. Most Favorable uses assumptions most
favorable to the project, Medium uses values considered most likely, and Least Favorable uses
values least favorable to the project. Table 7 summarizes the analysis.
Table 7 Analysis of Three Cases

Most Least
Data Input Favorable Medium Favorable
Generated Traffic Growth Rate (from Figure 3) L M H
Discount Rate 6% 6% 6%
Maximum Peak Vehicles Per Lane 2,200 2,000 1,800
Before Average Traffic Speed (km/hr) 40 50 60
After Average Traffic Speed (km/hr) 110 100 90
Value of Peak-Period Travel Time (per veh-hr) $12.00 $8.00 $6.00
Vehicle Operating Costs (per km) $0.15 $0.12 $0.10
Annual Lane Hours at Capacity Each Direction 1,200 1,000 800
Diverted Trip External Costs (per km) $0.00 $0.10 $0.15
Induced Travel External Costs (per km) $0.20 $0.30 $0.50
Net Present Value (millions)
NPV Without Consideration of Generated Traffic $204.8 $45.2 -$9.8
NPV With Consideration of Generated Traffic $124.5 -$32.1 -$95.7
Difference -$80.3 -$77.3 -$85.8
Benefit/Cost Ratio
Without Generated Traffic 6.90 2.30 0.72
With Generated Traffic 3.37 0.59 0.11
This table summarizes the assumptions used in this analysis.
21
The most favorable assumptions result in a positive B/C even when generated traffic is
considered. The medium assumptions result in a positive B/C if generated traffic is ignored but a
negative NPV if generated traffic is considered. The least favorable assumptions result in a
negative B/C even when generated traffic is ignored. In each case, considering generated traffic
has significant impacts on the results.
Figure 11 illustrates project benefits and costs based on “Medium” assumptions, ignoring
generated traffic. This results in a positive NPV of $45.2 million, implying that the project is
economically worthwhile.
Figure 11 Estimated Costs and Benefits, Ignoring Generated Traffic

Costs and Benefits
Vehicle Operating Cost Savings
Travel Time Savings
Project Costs
Years ==>
This figure illustrates annual benefits and costs when generated traffic is ignored, using
“Medium” assumptions. Benefits are bars above the baseline, costs are bars below the baseline.
Project expenses are the only cost category.
Figure 10 illustrates project evaluation when generated traffic is considered. Congestion

reduction benefits decline, and additional external costs and consumer benefits are included.
The NPV is –$32.1 million, indicating the project is not worthwhile.
Figure 10 Estimated Costs and Benefits, Considering Generated Traffic

Costs and Benefits
Generated Travel Benefits

Vehicle Operating Cost Savings
Travel Time Savings
Induced Travel External Costs
Diverted Trip External Costs
Years ==>
Project Expenses
This figure illustrates benefits and costs when generated traffic is considered, using medium
assumptions. Benefits are bars above the baseline, costs are bars below the baseline. It includes
consumer benefits and external costs associated with generated traffic. Travel time and vehicle
operating cost savings end after about 10 years, when traffic volumes per lane return to pre-
project levels, resulting in no congestion reduction benefits after that time.
22
This analysis indicates how generated traffic can have significant impacts on project assessment.
Ignoring generated traffic exaggerates the benefits of highway capacity expansion by
overestimating congestion reduction benefits and ignoring incremental external costs from
generated traffic. This tends to undervalue alternatives such as road pricing, TDM programs,
other modes, and “do nothing” options.
For example, Figure 11 compares three possible responses to congestion on a corridor with
increasing traffic demand. Do nothing causes traffic congestion costs to increase over time.
Expanding general traffic lanes imposes large initial costs due to construction delays, but
provides large short-term congestion reduction benefits. However, these decline over time, due
to induced traffic, and the additional vehicle travel imposes additional external costs including
downstream congestion, increased parking demand, accident risk and pollution emissions.
Building grade-separated public transit (either a bus lane or rail line) also imposes short-run
congestion delays, and the congestion reduction benefits are relatively small in the short term
but increase over time as transit ridership grows, networks expand, and development becomes
more transit-oriented.
Figure 11 Road Widening Versus Transit Congestion Impacts

Do N othing
Highway Expansion
Public Transit
Costs and Benefits
C onstructi on
Years = =>
A Do Nothing causes congestion costs to increase in the future. Highway expansion imposes
short term construction delays, then large congestion reduction benefits, but these decline over
time due to generated traffic. Grade-separated public transit provides smaller benefits in the
short-term but these increase over time as public transit ridership grows.
23
Counter Arguments
“Widening roads to ease congestion is like trying to cure obesity by loosening your belt” Roy
Kienitz, executive director of the Surface Transportation Policy Project
“Increasing highway capacity is equivalent to giving bigger shoes to growing children” Robert
Dunphy of the Urban Land Institute
Some highway expansion advocates claim that generated traffic has minor implications for
transport planning decisions. They argue that increased highway capacity contributes little to
overall growth in vehicle travel compared with other factors such as increased population,
employment and income (Heanue 1998; Burt and Hoover 2006), that although new highways
generate traffic, they still provide net economic benefits (ULI 1989), and that increasing roadway
capacity does reduce congestion (TRIP 1999; Bayliss 2008).
These arguments ignore critical issues, and are often based on outdated data and inaccurate
analysis. Overall travel trends indicate little about the cost effectiveness of particular policies
and projects. For example, studies which indicate that, in the past, increased lane-miles caused
minimal growth in vehicle travel (Burt and Hoover 2006), provide little guidance for future
planning, since, in the past, much of the added highway lane-miles occurred on uncongested
rural highways while most future highway expansion occurs on congested urban highways.
Strategies that encourage more efficient use of existing capacity, such as commute trip
reduction programs and road pricing, may provide greater social benefits, particularly
considering all costs (Goodwin 1997).
Highway expansion advocates generally ignore or severely understate generated traffic and
induced travel impacts. For example, Cox and Pisarski (2004) use a model that accounts for
diverted traffic (trips shifted in time or route) but ignores shifts in mode, destination and trip
frequency. Hartgen and Fields (2006) assume that generated traffic would fill just 15% of added
roadway capacity, based on generated traffic rates during the 1960s and 1970s, which is
unrealistically low when extremely congested roads are expanded. They ignore the incremental
costs that result from induced vehicle travel, such as increased downstream traffic congestion,
road and parking costs, accidents and pollution emissions. They claim that roadway capacity
expansion reduces fuel consumption, pollution emissions and accidents, because they measure
impacts per vehicle-mile and ignore increased vehicle miles. As a result they significantly
exaggerate roadway expansion benefits and understate total costs.
Debates over generated traffic and its implications often reflect ideological perspectives
concerning whether automobile travel (and therefore road capacity expansion) is “good” or
“bad”. To an economist, such arguments are silly. Some automobile travel provides large net
benefits (high user value, poor alternatives, low external costs), and some provides negative net
benefits (low user value, good alternatives, and large external costs). The efficient solution to
congestion is to use pricing or other incentives to test consumers’ willingness to pay for road
space and capacity expansion.
If consumers only demand roadway improvements when they are shielded from the true costs,
such projects are likely to be economically inefficient. Only if users are willing to pay the full
incremental costs their vehicle use imposes can society be sure that increased road capacity and
24
the additional vehicle travel that results provides net benefits. Travel demand predictions based
on underpriced roads overestimate the economically optimal level of roadway investments and
capacity expansion. Increasing capacity in such cases is more equivalent to loosening a belt than
giving a growing child larger shoes (see quotes above), since the additional vehicle travel is a
luxury and economically inefficient.
Some highway advocates suggest there are equity reasons to subsidize roadway capacity
expansion, to allow lower-income households access to more desirable locations, but most
benefits from increased roadway capacity are captured by middle- and upper-income
households (Deakin, et al. 2020). Improving travel choices for non-drivers tends to have greater
equity benefits than subsidizing additional highway capacity since physically and economically
disadvantaged people often rely on alternative modes.
Although highway projects are often justified for the sake of economic development, highway
capacity expansion now provides little net economic benefit (Boarnet 1997). An expert review
concluded, “The available evidence does not support arguments that new transport investment
in general has a major impact on economic growth in a country with an already well-developed
infrastructure” (SACTRA 1997). Melo, Graham and Canavan (2012) found a positive relationship
between U.S. urban highway expansion and economic output between 1982 and 2009, but no
reduction in long-term congestion. They conclude that other types of transportation system
improvements could provide greater economic development benefits.
25
Alternative Transport Improvement Strategies

Generated traffic significantly reduces roadway capacity expansion benefits, making other
congestion reduction solutions relatively more cost effective and beneficial. The article,
Spreading the Gospel of Induced Demand (Klein, et al. 2022) argues that the general public
misunderstands induced travel impacts and so tends to overestimate highway expansion
benefits and underestimate the benefits of alternative congestion solutions.
Considering generated traffic tends to increase the estimated value of improvements to

alternative modes (particularly grade-separated ridesharing and public transit services),
transportation systems management, efficient road pricing, and transportation demand
management strategies that result in more efficient use of existing roadway capacity. Although
these strategies may not necessarily eliminate traffic congestion individually, an integrated
package can significantly reduce congestion delays compared with what would otherwise occur,
usually with less costs and greater total benefits than highway capacity expansion. Below are
examples (VTPI 2001):
 Congestion pricing can provide travelers with an incentive to reduce their peak period trips
and use space-efficient modes such as public transit, ridesharing, bicycling and walking.
 Commute trip reduction programs can provide a framework for encouraging commuters to
drive less and rely more on travel alternatives.
 Land use management can increase access by bringing closer common destinations.
 Pedestrian and cycle improvements can increase mobility and access, and support other
modes such as public transit (since transit users also depend on walking and cycling).
 Public transit service that offers door-to-door travel times and user costs that are
competitive with driving can attract travelers from a parallel highway, limiting the
magnitude of traffic congestion on that corridor.
26
Legal Issues
Environmental groups successfully sued the Illinois transportation agencies for failing to
consider land use impacts and generated traffic in the Environmental Impact Statement (EIS) for
I-355, a proposed highway extension outside the city of Chicago (Sierra Club 1997). The federal
court concluded that the EIS was based on the “implausible” assumption that population in the
rural areas would grow by the same amount with and without the tollroad, even though project
was promoted as a way to stimulate growth. The court concluded that this circular reasoning
afflicted the document’s core findings. The judge required the agencies to prepare studies
identifying the amount of development the tollroad would cause, and compare this with
alternatives. The Court’s order states:
Plaintiffs’ argument is persuasive. Highways create demand for travel and expansion by their
very existence…Environmental laws are not arbitrary hoops through which government
agencies must jump. The environmental regulations at issue in this case are designed to
ensure that the public and government agencies are well informed about the environmental
consequences of proposed actions. The environmental impact statements in this case fail in
several significant respects to serve this purpose. (ELCP)
In 2008 the California Attorney General recognized that regional transportation plans must
consider induced travel impacts when evaluating the climate change impacts of individual
projects to meet California Environmental Quality Act (CEQA) requirements (Brown 2008). CEQA
requires that “[e]ach public agency shall mitigate or avoid the significant effects on the
environment of projects that it carries out or approves whenever it is feasible to do so.” The
state Attorney General recognizes that transportation planning decisions, such as highway
expansion projects, can have significant emission impacts due to induced vehicle travel.
Some new laws and regulations, such as California Senate Bill 743 (S.B. 743), prohibit the use of
vehicle level of service (LOS) and similar measures as the sole basis for evaluating transportation
improvement options; instead, policies and project are evaluated based on their ability to
reduce vehicle miles traveled (VMT). This will require consideration of induced travel effects in
analysis of roadway projects (Milam, et al. 2017).
In 2020, the California Department of Transportation established specific requirements for

evaluating and mitigating the induced travel impacts of roadway expansion projects (Sundquist
2020), based on an extensive expert review (Caltrans 2020a). This analysis is based on lane-
miles-to-induced-VMT elasticities, as specified in the Transportation Analysis Framework:
Induced Travel Analysis report (Caltrans 2020b), and estimated by the National Center for
Sustainable Transportation’s Induced Travel Calculator (NCST 2019).
27
Conclusions
Urban traffic congestion tends to maintain equilibrium. Congestion reaches a point at which it
discourages additional peak-period trips. Increasing road capacity allows more vehicle travel to
occur. In the short term this consists primarily of generated traffic: vehicle travel diverted from
other times, modes, routes and destinations. Over the long run an increasing portion consists of
induced vehicle travel, resulting in a total increase in regional VMT. This has several implications
for transport planning:
 Ignoring generated traffic underestimates the magnitude of future traffic congestion problems,
overestimates the congestion reduction benefits of increasing roadway capacity, and
underestimates the benefits of alternative solutions to transportation problems.
 Induced travel increases many external costs. Over the long term it helps create more
automobile dependent transportation systems and land use patterns.
 The mobility benefits of generated traffic are relatively small since they consist of marginal value
trips. Much of the benefits are often capitalized into land values.
Ignoring generated traffic results in self-fulfilling predict and provide planning: Planners
extrapolate traffic growth rates to predict that congestion will reach gridlock unless capacity
expands. Adding capacity generates traffic, which leads to renewed congestion with higher
traffic volumes, and more automobile oriented transport and land use patterns. This cycle
continues until road capacity expansion costs become unacceptable.
The amount of traffic generated depends on specific conditions. Expanding highly congested
roads with considerable latent demand tends to generate significant amounts of traffic,
providing only temporary congestion reductions.
Generated traffic does not mean that roadway expansion provides no benefits and should never
be implemented. However, ignoring generated traffic results in inaccurate forecasts of impacts
and benefits. Road projects considered cost effective by conventional analysis may actually
provide little long-term benefit to motorists and make society overall worse off due to induced
travel external costs. Other strategies may be better overall. Another implication is that highway
capacity expansion projects should incorporate strategies to avoid increasing external costs,
such as more stringent vehicle emission regulations to avoid increasing pollution and land use
regulations to limit sprawl.
Framing the Congestion Question

If you ask people, “Do you think that traffic congestion is a serious problem?” they frequently answer yes. If
you ask, “Would you rather solve congestion problems by improving roads or by using alternatives such as
congestion tolls and other TDM strategies?” a smaller majority would probably choose the road
improvement option. This is how transport choices are generally framed.
But if you present the choices more realistically by asking, “Would you rather spend a lot of money to
increase road capacity to achieve moderate and temporary congestion reductions and bear higher future
costs from increased motor vehicle traffic, or implement other types of transportation improvements?” the
preference for road building is likely to decline.
28
References and Information Resources
ADB (2010), Reducing Carbon Emissions from Transport Projects, Asian Development Bank
(www.adb.org); at www.adb.org/evaluation/reports/ekb-carbon-emissions-transport.asp.
Asif Ahmed and Peter Stopher (2014), “Seventy Minutes Plus or Minus 10 — A Review of Travel
Time Budget Studies,” Transport Reviews, Vo. 34:5, pp. 607-625
(DOI: 10.1080/01441647.2014.946460).
Anupriya, Prateek Bansal and Daniel J. Graham (2023), “Congestion in Cities: Can Road Capacity
Expansions Provide a Solution?,” Transportation Research Part A, Vo. 174
(https://doi.org/10.1016/j.tra.2023.103726).
Richard Arnott and Kenneth Small (1994), “The Economics of Traffic Congestion,” American
Scientist, Vol. 82, Sept./ Oct. 1994, pp. 446-455.
Lawrence C. Barr (2000), “Testing for the Significance of Induced Highway Travel Demand in
Metropolitan Areas,” Transportation Research Record 1706, Transportation Research Board
(www.trb.org); at http://trb.metapress.com/content/lq766w66540p7432/fulltext.pdf.
Matthew Barth and Kanok Boriboonsomsin (2009), “Traffic Congestion and Greenhouse Gases,”
Access 35, (www.accessmagazine.org); at https://bit.ly/3nwZIjb.
David Bayliss (2008), Misconceptions and Exaggerations about Roads and Road Building in Great
Britain, Royal Automobile Club Foundation
(www.racfoundation.org/index.php?option=com_content&task=view&id=597&Itemid=35)
Marlon Boarnet (1997), Direct and Indirect Economic Effects of Transportation Infrastructure,
UCTC (www.uctc.net).
Marlon Boarnet and Saksith Tan Chalermpong (2002), New Highways, Induced Travel and Urban
Growth Patterns: A "Before and After" Test, Paper 559, University of California Transportation
Center (www.uctc.net).
Robert Burchell, et al. (1998), Costs of Sprawl – Revisited, TCRP Report 39, Transportation
Research Board (www.trb.org).
Michael Burt and Greg Hoover (2006), Build It and Will They Drive? Modelling Light-Duty Vehicle
Travel Demand, Conference Board of Canada (www.conferenceboard.ca); at https://bit.ly/39IxQqZ.
Edward G. Brown (2008), Comments on the Notice of Preparation for Draft Environmental
Impact Report for the Transportation 2035 Plan, California Attorney General (http://ag.ca.gov);
at http://ag.ca.gov/globalwarming/pdf/comments_MTC_RT_Plan.pdf.
Jihye Byun, Sungjin Park and Kitae Jang (2017), “Rebound Effect or Induced Demand? Analyzing
the Compound Dual Effects on VMT in the U.S.,” Sustainability, (www.mdpi.com/2071-
1050/9/2/219/pdf).
29
Sally Cairns, C. Hass-Klau and Phil Goodwin (1998), Traffic Impacts of Highway Capacity
Reductions: Assessment of the Evidence, London Transport Planning; at https://bit.ly/3ME2aS3.
Caltrans (2020), Transportation Analysis Framework: Induced Travel Analysis, California

Department of Transportation (https://dot.ca.gov); at https://bit.ly/3aCYo7M.
Alison Cassady, Tony Dutzik and Emily Figdor (2004), More Highways, More Pollution: Road-
Building and Air Pollution in American's Cities, U.S. PIRG Education Fund (www.uspirg.org).
Robert Cervero (2002), “Induced Travel Demand: Research Design, Empirical Evidence, and
Normative Policies,” Journal of Planning Literature, Vol. 17, No. 1, pp. 4-19
Robert Cervero (2003a), “Are Induced Travel Studies Inducing Bad Investments?,” ACCESS,
Number 22, University of California Transportation Center (www.uctc.net), Spring, 22-27.
Robert Cervero (2003b), “Road Expansion, Urban Growth, and Induced Travel: A Path Analysis,”
Journal of the American Planning Association, Vol. 69, No. 2 (www.planning.org), pp. 145-163.
Robert Cervero and Mark Hansen (2000), Road Supply-Demand Relationships: Sorting Out
Casual Linkages, Institute of Transportation Studies, University of California (www.uctc.net).
CNU (2012), Highways to Boulevards Program, Congress for New Urbanism

(www.cnu.org/highways).
Wendell Cox and Alan Pisarski (2004), Blueprint 2030: Affordable Mobility And Access For All,
Georgians for Better Mobility (http://ciprg.com/ul/gbt/atl-report-20040621.pdf).
J. M. Dargay and P. B. Goodwin (1995), “Evaluation of Consumer Surplus with Dynamic Demand
Changes.” Journal of Transport Economics and Policy, Vol. XXIX, No. 2, pp. 179-93.
Elizabeth Deakin, et al. (2020), Calculating and Forecasting Induced Vehicle Miles of Travel
Resulting from Highway Projects: Findings and Recommendations from an Expert Panel,
California Department of Transportation (https://dot.ca.gov); at https://bit.ly/3nwaf10.
Patrick DeCorla-Souza (2000), “Estimating Highway Mobility Benefits,” ITE Journal

(www.ite.org), February 2000, pp. 38-43. Also, Patrick DeCorla-Souza, “Evaluating the Trade-Offs
Between Mobility and Air Quality,” ITE Journal, February 2000, pp. 65-70.
Patrick DeCorla Souza and Harry Cohen (1999), “Estimating Induced Travel For Evaluation of
Metropolitan Highway Expansion,” Transportation, Vol. 26, pp. 249-261.
Mark Delucchi (1996), “Total Cost of Motor-Vehicle Use,” Access

(http://violet.berkeley.edu/~uctc), No. 8, Spring 1996, pp. 7-13.
DfT (2007), Transport Analysis Guidance, UK Department For Transport (www.webtag.org.uk).

The Variable Demand Modelling (www.dft.gov.uk/webtag/documents/expert/unit3.10.1.php)
section describes methods for incorporating induced travel demand into project appraisal.
30
Anthony Downs (1992), Stuck in Traffic, Brookings Institution (www.brookings.edu); at

www.brookings.edu/book/stuck-in-traffic.
Gilles Duranton and Matthew A. Turner (2011), “The Fundamental Law of Highway Congestion:
Evidence from the US,” American Economic Review (www.aeaweb.org); at
www.nber.org/papers/w15376 and https://bit.ly/35H0jJk.
Mary Ebeling (2014), Re-Thinking the Urban Freeway, State Smart Transportation Initiative
(https://ssti.us) and Mayors Innovation Project; at https://bit.ly/3mLriZG.
EC (2004), Reclaiming City Streets for People: Chaos or Quality Of Life? European Commission
(www.ec.europa.eu); at http://ec.europa.eu/environment/pubs/pdf/streets_people.pdf.
Reid Ewing and Shima Hamidi (2014), Measuring Urban Sprawl and Validating Sprawl Measures,
Metropolitan Research Center at the University of Utah for the National Cancer Institute, the
Brookings Institution and Smart Growth America (www.smartgrowthamerica.org); at
https://bit.ly/2I6StdG.
FHWA (1998), Surface Transportation Efficiency Analysis Model (STEAM), Federal Highway
Administration (www.fhwa.dot.gov/steam).
FHWA (1999), Social Costs of Alternative Land Development Scenarios, Federal Highway
Administration (www.fhwa.dot.gov/scalds/scalds.html).
FHWA (2002), Highway Economic Requirements System: Technical Report, FHWA, USDOT
(https://trid.trb.org/view/725673); at https://rosap.ntl.bts.gov/view/dot/58541.
FHWA (2004), Induced Travel: Frequently Asked Questions, Federal Highway Administration
(www.fhwa.dot.gov/planning/itfaq.htm).
Georgios Fontaras, et al. (2014), “Development and Review of Euro 5 Passenger Car Emission
Factors Based on Experimental Results over Various Driving Cycles,” Science of the Total
Environment, Vo. 468–469, Pages 1034-1042 (https://doi.org/10.1016/j.scitotenv.2013.09.043).
Miquel-Àngel Garcia-López, Ilias Pasidis, and Elisabet Viladecans-Marsal (2020), Congestion in

Highways when Tolls and Railroads Matter: Evidence from European Cities, Universitat
Autonoma of Barcelona (https://ecap.uab.cat/RePEc/doc/wpdea2011.pdf), Ideas
(https://ideas.repec.org); at https://ideas.repec.org/p/uab/wprdea/wpdea2011.html.
Phil Goodwin (1996), “Empirical Evidence on Induced Traffic,” Transportation, Vo. 23, No. 1, pp.
35-54 (https://doi.org/10.1007/BF00166218).
Phil Goodwin (1997), Solving Congestion, Inaugural lecture for the Professorship of Transport
Policy, University College London; at http://discovery.ucl.ac.uk/1244/1/2004_22.pdf.
Phil B. Goodwin (1998), “The End of Equilibrium,” in Theoretical Foundations of Travel Choice
Modelling. edited by T. Garling et al., Elsevier, ISBN 0080430627.
31
Phil Goodwin (2002), “Disappearing Traffic? The Story So Far,” Proceedings of the Institution of
Civil Engineers; Municipal Engineer, Vo. 151, Issue 1 (www.municipalengineer.com) March 2002,
pp. 13-22; at www.onestreet.org/images/stories/Disappearing_traffic.pdf.
Phil Goodwin and Robert B. Noland (2003), “Building New Roads Really Does Create Extra
Traffic: A Response to Prakash et al.,” Applied Economics (www.tandf.co.uk/journals).
Phil Goodwin and Lisa Hopkinson (2023), Induced Traffic: Yet Again a Worryingly Overlooked
Dimension in Crucial Road Planning and Appraisal Policy, TAPAS.network
(https://tapas.network); at https://tapas.network/35/hopkinsongoodwin.php.
Roger Gorham (2009), Demystifying Induced Travel Demand, Sustainable Transportation

Technical Document, Sustainable Urban Transportation Project (www.sutp.org) and GTZ
(www.gtz.de); at http://tinyurl.com/l5q4ktp.
Dan J. Graham, Emma J. McCoy and David A. Stephens (2014), “Quantifying Causal Effects of
Road Network Capacity Expansions on Traffic Volume and Density via a Mixed Model Propensity
Score Estimator,” Journal of the American Statistical Association, Vol. 109, pp. 1440-1449,
(http://dx.doi.org/10.1080/01621459.2014.956871); earlier version at https://bit.ly/2CtzfBw.
Susan Handy and Marlon G. Boarnet (2014), Impact of Highway Capacity and Induced Travel on
Passenger Vehicle Use and Greenhouse Gas Emissions; Technical Background Document,
California Air Resources Board (www.arb.ca.gov); at https://bit.ly/2O4exJA.
Susan Handy (2015), Increasing Highway Capacity Unlikely to Relieve Traffic Congestion, National
Center for Sustainable Transportation (http://ncst.ucdavis.edu); at https://bit.ly/1QlDn4z.
Mark Hansen, et al. (1993), Air Quality Impacts of Urban Highway Capacity Expansion: Traffic
Generation and Land Use Changes, Institute of Transport Studies (www.uctc.net); at
https://escholarship.org/uc/item/6zz3k76c.
Mark Hansen (1995), “Do New Highways Generate Traffic?” Access No. 7 (www.uctc.net), Fall,
pp.16-22; at www.accessmagazine.org/fall-1995/do-new-highways-generate-traffic.
Mark Hansen and Yuanlin Huang (1997), “Road Supply and Traffic in California Urban Areas,”
Transportation Research A, Vol. 31, No. 3, pp. 205-218; at http://bit.ly/2A0Gi1C.
David T. Hartgen (2003), The Impact of Highways and Other Major Road Improvements on Urban
Growth in Ohio, The Buckeye Institute (www.buckeyeinstitute.org).
David T. Hartgen and M. Gregory Fields (2006), Building Roads to Reduce Traffic Congestion in
America’s Cities: How Much and at What Cost? Reason Foundation (www.reason.org).
Greig Harvey and Elizabeth Deakin (1993), A Manual of Regional Transportation Modeling
Practice for Air Quality, National Association of Regional Councils (www.narc.org).
Kevin Heanue (1998), Highway Capacity Expansion and Induced Travel; Evidence and
Implications, Circular 481, TRB (www.trb.org); at https://bit.ly/3IEgF77.
32
Peter Hills (1996), “What is Induced Traffic?” Transportation, Vol. 23, No. 1, pp. 5-16.
Kent M. Hymel, Kenneth A. Small and Kurt Van Dender (2010), “Induced Demand and Rebound
Effects in Road Transport,” Transportation Research B (www.elsevier.com/locate/trb), Vol.
44/10, Dec., pp. 1220-1241; at www.socsci.uci.edu/~ksmall/Rebound_congestion_27.pdf.
Kent Hymel (2019), “If You Build it, They Will Drive: Measuring Induced Demand for Vehicle Travel
in Urban Areas,” Transport Policy, Vo. 76, pp. 57-66 (doi.org/10.1016/j.tranpol.2018.12.006).
ICF Consulting (2005), Handbook on Integrating Land Use Considerations into Transportation
Projects to Address Induced Growth, AASHTO; at
https://onlinepubs.trb.org/onlinepubs/archive/notesdocs/25-25(3)_fr.pdf.
Induced Travel Calculator (https://blinktag.com/induced-travel-calculator), by the University of

California Davis National Center for Sustainable Transportation, estimates the additional annual
vehicle miles of travel (VMT) induced by state highway expansions.
ITDP (2012), Death and Life of Urban Highways, Institute for Transportation and Development
Policy (www.itdp.org); at www.itdp.org/2012/03/13/the-life-and-death-of-urban-highways.
ITF (2021), Reversing Car Dependency, International Transport Forum (www.itf-oecd.org); at

www.itf-oecd.org/sites/default/files/docs/reversing-car-dependency.pdf.
Peter Jacobsen (1997), “Liquid vs. Gas Models for Traffic,” Los Angeles Times, Letters to Editor,
14 May 1997; at www.latimes.com/archives/la-xpm-1997-05-14-me-58478-story.html.
Annika K. Jägerbrand, et al. (2014), Rebound Effects of Energy Efficiency Measures in the
Transport Sector in Sweden, The Swedish Energy Agency (www.vti.se); at https://bit.ly/3zgzfQ0.
Robert Johnston and Raju Ceerla (1996), “The Effects of New High-Occupancy Vehicle Lanes on
Travel and Emissions,” Transportation Research, Vol. 30A, No. 1, pp. 35-50.
Robert A. Johnston, et al. (2001), Applying an Integrated Model to the Evaluation of Travel
Demand Management Policies in the Sacramento Region, Mineta Transportation Institute
(www.transweb.sjsu.edu); at http://transweb.sjsu.edu/publications/01-03.pdf.
Nicholas J. Klein, et al. (2022), “Spreading the Gospel of Induced Demand,” Transfer Magazine,
Issue 9 (https://transfersmagazine.org); at https://bit.ly/3arJWFj.
Kara Kockelman (2011), “Traffic Congestion,” Chapter 22, Transportation Engineering Handbook,
McGraw Hill (www.mhprofessional.com); pre-print at https://bit.ly/3PyX3Er.
Robert Krol (2020), Can We Build Our Way Out of Urban Traffic Congestion?, Center for Growth
and Opportunity at Utah State University (www.thecgo.org); at https://bit.ly/2ZvbDUv.
Douglass Lee, Lisa Klein and Gregorio Camus (1999), “Induced Traffic and Induced Demand,”
Transportation Research Record 1659, TRB (www.trb.org), pp. 68-75 (doi.org/10.3141/1659-09).
33
David Levinson and Ajay Kumar (1995), “Activity, Travel, and the Allocation of Time,” APA
Journal, Vol. 61, No. 4, pp. 458-470 (https://doi.org/10.1080/01944369508975657). .
Todd Litman (2001a), What’s It Worth? Life Cycle and Benefit/Cost Analysis for Evaluating
Economic Value, Presented at Internet Symposium on Benefit-Cost Analysis, Transportation
Association of Canada (www.tac-atc.ca); at www.vtpi.org/worth.pdf.
Todd Litman (2001), “Generated Traffic: Implications for Transport Planning,” ITE Journal, Vo.
71/4, Institute of Transportation Engineers (www.ite.org), pp. 38-47; at www.vtpi.org/gentraf.pdf.
Todd Litman (2003), Transportation Cost Analysis: Techniques, Estimates and Implications,
Victoria Transport Policy Institute (www.vtpi.org/tca).
Todd Litman (2009), Transportation Elasticities, Victoria Transport Policy Institute

(www.vtpi.org); at www.vtpi.org/elasticities.pdf.
Todd Litman (2012), “Changing North American Vehicle-Travel Price Sensitivities: Implications
for Transport and Energy Policy” Transport Policy
(http://dx.doi.org/10.1016/j.tranpol.2012.06.010); at www.vtpi.org/VMT_Elasticities.pdf.
Todd Litman (2013), Transport Elasticities: Impacts on Travel Behaviour: Understanding

Transport Demand To Support Sustainable Travel Behavior, Technical Document #11,
Sustainable Urban Transport Project (www.sutp.org) and GIZ (www.giz.de); at
https://sutp.org/publications/transport-elasticities-impacts-on-travel-behaviour.
Todd Litman (2014), Congestion Evaluation Best Practices, International Transportation

Economic Development Conference, Dallas, Texas; at www.vtpi.org/ITED_congestion.pdf.
Todd Litman (2014b), Congestion Costing Critique: Critical Evaluation of the ‘Urban Mobility
Report,’ Victoria Transport Policy Institute (www.vtpi.org); at www.vtpi.org/UMR_critique.pdf.
Todd Litman (2016), Transportation Land Use Impacts, Victoria Transport Policy Institute
(www.vtpi.org); at www.vtpi.org/landuse.pdf.
Todd Litman (2019), Smart Congestion Relief: Comprehensive Analysis of Traffic Congestion
Costs and Congestion Reduction Benefits, Victoria Transport Policy Institute (www.vtpi.org); at
www.vtpi.org/cong_relief.pdf.
Todd Litman (2021), Not So Fast: Better Speed Valuation for Transportation Planning, Victoria
Transport Policy Institute (www.vtpi.org); at https://vtpi.org/nsf.pdf.
William R. Loudon, Janaki Parameswaran and Brian Gardner (1997), “Incorporating Feedback in
Travel Forecasting: Methods, Pitfalls and Common Concerns,” Transportation Research Record
1607, TRB (www.trb.org), pp. 185-195; at http://trb.metapress.com/content/120399.
Peter Mackie (1996), “Induced Traffic and Economic Appraisal,” Transportation, Vol. 23
(https://doi.org/10.1007/BF00166222).
34
Cesare Marchetti (1994), “Anthropological Invariants in Travel Behaviour, Technological

Forecasting and Social Change,” p. 75-88; at https://bit.ly/2sYBJjR. Also see Jesse H. Ausubel and
Cesare Marchetti (2001), “Evolution of Transport,” Industrial Physicist; at https://bit.ly/3v1IgZz.
Norman Marshall (2016), A Statistical Model of Regional Traffic Congestion in the United States,
TRB Annual Meeting (https://trid.trb.org/view/1392295); at https://bit.ly/3GNI2L1.
METRO (2022), Climate Emissions Analysis: Metro’s Indirect Impact on Greenhouse Gas
Emissions, Los Angeles Region METRO (www.metro.net); at https://bit.ly/3B5QGDo.
David Metz (2021), “Economic Benefits of Road Widening: Discrepancy Between Outturn and
Forecast,” Transportation Research Part A, Vo. 147, pp. 312-319
(https://doi.org/10.1016/j.tra.2021.03.023).
Ronald T. Milam, et al. (2017), “Closing the Induced Vehicle Travel Gap Between Research and
Practice,” Transportation Research Record 2653 (https://doi.org/10.3141/2653-02).
Martin Mogridge (1997), “The Self-Defeating Nature of Urban Road Capacity Policy; A Review of
Theories, Disputes and Available Evidence,” Transport Policy, Vo. 4, No. 1, pp. 5-23; at
https://bit.ly/39F3Try. Also see Lewis-Mogridge Position, Wikipedia
(http://en.wikipedia.org/wiki/Lewis-Mogridge_Position).
Patricia C. Melo, Daniel J. Graham and Shane Canavan (2012), “Effects of Road Investments on
Economic Output and Induced Travel Demand,” Transportation Research Record 2297,
Transportation Research Board (www.trb.org), pp. 163-171; at http://tinyurl.com/mwdvfsl.
Petter Næss, Morten Skou Nicolaisen and Arvid Strand (2012), “Traffic Forecasts Ignoring
Induced Demand: a Shaky Fundament for Cost-Benefit Analyses,” European Journal of Transport
and Infrastructure Research, Vol. 12 (3), pp. 291-301; at https://bit.ly/2OR84SC.
NCST (2019), Induced Travel Calculator, National Center for Sustainable Transportation
(https://ncst.ucdavis.edu); at https://blinktag.com/induced-travel-calculator.
Peter Newman and Jeff Kenworthy (1998), Sustainability and Cities; Overcoming Automobile
Dependency, Island Press (www.islandpress.org).
Robert Noland (2001), “Relationships Between Highway Capacity and Induced Vehicle Travel,”
Transportation Research A, Vol. 35, No. 1, January 2001, pp. 47-72.
Robert B. Noland and Lewison L. Lem (2002), “A Review of the Evidence for Induced Travel and
Changes in Transportation and Environmental Policy in the US and the UK,” Transportation
Research D, Vol. 7, No. 1 (www.elsevier.com/locate/trd), pp. 1-26.
Robert Noland and Mohammed A. Quddus (2006), “Flow Improvements and Vehicle Emissions:
Effects of Trip Generation and Emission Control Technology,” Transportation Research D, Vo. 11
(www.elsevier.com/locate/trd), pp. 1-14; at https://bit.ly/3kueGX5.
35
Robert B. Noland and Christopher S. Hanson (2013), “How Does Induced Travel Affect
Sustainable Transport Policy,” Transportation Beyond Oil: Policy Choices for a Multimodal
Future, (John Renne and Billy Fields, eds), Island Press (www.islandpress.com), pp. 70-85.
John Odgers (2009), Have All The Travel Time Savings On Melbourne’ Road Network Been
Achieved?, A GAMUT Discussion Paper, School of Management, RMIT University; at
www.abp.unimelb.edu.au/gamut/pdf/have-all-the-time-savings-been-achieved.pdf.
M. Özuysal and S. Tanyel (2008), “Induced Travel Demand in Developing Countries: Study of
State Highways in Turkey,” Journal of Urban Planning and Development, Vo. 134, No 2, pp. 78-
87; abstract at http://ascelibrary.org/upo/resource/1/jupddm/v134/i2/p78_s1.
Mohammad Lutfur Rahmana, Douglas Bakerb and M. Shafiq-Ur Rahmanc (2020), “Modelling
Induced Travel Demand in a Developing Country: Evidence from Dhaka, Bangladesh,”
Transportation Research Procedia 48, pp. 3439–3456 (10.1016/j.trpro.2020.08.108); at
www.sciencedirect.com/science/article/pii/S235214652030524X.
Stuart Ramsey (2005), “Of Mice and Elephants,” ITE Journal, Vol. 75, No. 9 (www.ite.org), pp. 38-
41; at https://bit.ly/2MuCKfX.
Caroline Rodier, et al. (2001), Anatomy of Induced Travel; Using an Integrated Land Use and
Transportation Model in the Sacramento Region, Transportation Research Board Annual
Meeting (www.trb.org).
Peter Romilly (2004), “Welfare Evaluation With a Road Capacity Constraint,” Transportation
Research A, Vol. 38/4 (DOI: 10.1016/j.tra.2003.12.001), pp. 287-303; at https://bit.ly/3lwiNmW.
SACTRA (1994), Trunk Roads and the Generation of Traffic, Standing Advisory Committee on
Trunk Road Assessment, UKDoT, HMSO (London; www.roads.detr.gov.uk/roadnetwork).
SACTRA (1997), Transport Investment, Transport Intensity and Economic Growth: Interim Report,
Standing Committee on Trunk Road Assessment, Dept. of Environment, Transport and Regions
(www.roads.detr.gov.uk/roadnetwork/heta/sactra98.htm).
Robert G. Schiffer, M. Walter Steinvorth and Ronald T. Milam (2005), Comparative Evaluations
on the Elasticity of Travel Demand, TRB Paper 05-0313 (www.trb.org); at https://bit.ly/3Q1LZzU.
Benjamin Schneider (2018), Induced Demand, CityLab University (www.bloomberg.com/citylab);

at https://bloom.bg/2VDYfwd.
SHIFT Calculator (https://shift.rmi.org). The State Highway Induced Frequency of Travel (SHIFT),
developed by the Rocky Mountain Institute, estimates long-run (i.e., after 5 to 10 years) induced
vehicle travel and emissions impacts from highway expansions in U.S. urban regions. It is based
on methodologies developed for California’s Transportation Analysis Framework.
Sierra Club (1997), Ill. Chapter v U.S. Dept. of Transp. (962 F supp. 1037, ND Ill 1997).
36
STPP (1998), Do New Roads Cause Congestion?, Surface Transportation Policy Project
(www.transact.org); at www.transact.org/congestion/analysis.htm.
Eric Sundquist (2020), California Highway Projects Face Review for Induced Travel, State Smart
Transportation Initiative (www.ssti.us); at https://bit.ly/2VW5pev.
TfA (2020), The Congestion Con: How More Lanes and More Money Equals More Traffic,
Transportation for America (http://t4america.org); at http://t4america.org/wp-
content/uploads/2020/03/Congestion-Report-2020-FINAL.pdf.
Aud Tennøy, Anders Tønnesen and Frants Gundersen (2019), “Effects of Urban Road Capacity
Expansion – Experiences from Two Norwegian Cases,” Transportation Research Part D: Transport
and Environment, Vol. 69, pp. 90-106 (https://doi.org/10.1016/j.trd.2019.01.024).
TØI (2009), Does Road Improvement Decrease Greenhouse Gas Emissions? Norwegian Centre for
Transport Research (www.toi.no); summary at https://bit.ly/3HhHM6x.
Transportation, Vol. 23, No. 1, February 1996. Special issue devoted to induced travel.
TRB (1995), Expanding Metropolitan Highways: Implications for Air Quality and Energy Use,
Committee for Study of Impacts of Highway Capacity Improvements on Air Quality and Energy
Consumption, Transportation Research Board, Special Report #345 (www.trb.org).
TRIP (1999), The Best Solutions to Traffic Congestion: Dispelling the Myths about the Impact of
Expanding Roads, The Road Information Program (www.tripnet.org).
TRISP (2005), “Treatment of Induced Traffic,” Economic Evaluation Notes, UK Department for
International Development and the World Bank (www.worldbank.org); at
http://go.worldbank.org/ME49C4XOH0.
UKERC (2007), 'Rebound Effects' Threaten Success of UK Climate Policy, UK Energy Research Centre
(www.ukerc.ac.uk); at www.ukerc.ac.uk/support/tiki-index.php?page=0710ReboundEffects.
UKERC (2009), What Policies Are Effective At Reducing Carbon Emissions From Surface Passenger
Transport? UK Energy Research Centre; at
www.ukerc.ac.uk/Downloads/PDF/T/TPA_transport_final.pdf.
ULI (1989), Myths and Facts about Transportation and Growth, Urban Land Ins. (www.uli.org).
USEPA (2001), Our Built and Natural Environments: A Technical Review of the Interactions
Between Land Use, Transportation and Environmental Quality, US Environmental Protection
Agency (www.epa.gov/smartgrowth/pdf/built.pdf).
Han van der Loop (2014), The Latent Demand in Road Traffic, KiM Netherlands Institute for
Transport Policy Analysis (http://kimnet.nl); at https://bit.ly/2Y3icON.
William S. Vickrey (1969), “Congestion Theory and Transport Investment,” American Economic
Review, Vol. 59/2, May, pp. 251-260; at https://bit.ly/2S0pkrB.
37
Jamey M. B. Volker, Amy E. Lee and Susan Handy (2020), “Induced Vehicle Travel in the Environmental
Review Process,” Transportation Research Record (doi.org/10.1177/0361198120923365). Also see,
Jamey, Volker and Handy (2021), The Induced Travel Calculator and its Applications, University of
California Institute of Transportation Studies (http://dx.doi.org/10.7922/G22F7KQH).
Wanle Wang, Ming Zhong and John Douglas Hunt (2019), “Analysis of the Wider Economic Impact of a
Transport Infrastructure Project Using an Integrated Land Use Transport Model,” Sustainability
11(2):364 (DOI: 10.3390/su11020364).
Huw C. W. L. Williams and Yaeko Yamashita (1992), “Travel Demand Forecasts and the
Evaluation of Highway Schemes Under Congested Conditions,” Journal of Transport Economics
and Policy, Vol. 26, No. 3, pp. 261-282 (www.jstor.org/stable/20052987).
Clark Williams-Derry (2007), Increases in Greenhouse-Gas Emissions from Highway-Widening

Projects, Sightline Institute (www.sightline.org); at https://bit.ly/3pJ3haw.
WSP (2018), Latest Evidence on Induced Travel Demand: An Evidence Review, UK Department for
Transport (www.gov.uk); at https://bit.ly/3eeCm0r.
Enjian Yao and Takayuki Morikawa (2005), “A Study of An Integrated Intercity Travel Demand
Model,” Transportation Research A, Vol. 39, No. 4 (www.elsevier.com/locate/tra), May, pp 367-381.
Hyejin Youn, Hawoong Jeong and Michael T. Gastner (2008), “The Price of Anarchy in
Transportation Networks: Efficiency and Optimality Control,” Journal of Physical Letters; at
http://arxiv.org/PS_cache/arxiv/pdf/0712/0712.1598v4.pdf and https://bit.ly/3vRUZOJ.
David Zipper (2021), “The Unstoppable Appeal of Highway Expansion,” Bloomberg

(www.bloomberg.com); at www.bloomberg.com/news/features/2021-09-28/why-widening-
highways-doesn-t-bring-traffic-relief.
www.vtpi.org/gentraf.pdf
38

Gentraf

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Gentraf

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gentraf

Uploaded by

Copyright:

Available Formats

www.vtpi.

Generated Traffic and Induced Travel

Todd Litman  1998-2023

Defining Generated Traffic and Induced Vehicle Travel

Below are examples of decisions that generate traffic:

Figure 1 How Road Capacity Expansion Generates Traffic

Table 1 Types of Generated Traffic

Measuring Generated Traffic and Induced Vehicle Travel

Figure 4 Istanbul Bridge Traffic Volumes (Personal correspondence with

This graph shows Average

Figure 5 Elasticity of Traffic Volume With Respect to Road Capacity

Modeling Generated Traffic and Induced Travel

Short Cut Methods of Incorporating Induced Demand

Land Use Impacts

Costs of Induced Travel

Table 3 Motor Vehicle Indirect and External Costs (Delucchi 1996)

Table 4 Cost Impacts of Roadway Capacity Expansion

 Per-mile crash rates (if  Traffic services

Table 5 Estimated Incremental External Costs of Generated Traffic

“Road construction, largely speaking, increases greenhouse gas emissions, mainly

Table 6 summarizes roadway improvement emission impacts, including effects on emission

Table 6 Roadway Expansion Greenhouse Gas Emission Impacts (TØI 2009)

Calculating Consumer Benefits

Figure 6 Vehicle Travel Demand Curve Illustrating the Rule-of-Half

Reduced User Cost

Explanation of the “Rule of Half”

Figure 6 Speed-Emission Curves (Fontaras, et al. 2014)

Kilometers per Hour

Figure 7 U.S. Speed-Emission Curves (Barth and Boriboonsomsin 2009)

Figure 9 Projected Traffic

3,000 3 Lanes, Considering Generated Traffic

2,000 3 Lanes, Ignoring Generated Traffic

Figure 10 Projected Average Traffic Speeds

Av erage Traffic Speed (km/hr)

20 Considering Generated Traffic

Table 7 Analysis of Three Cases

Figure 11 Estimated Costs and Benefits, Ignoring Generated Traffic

Vehicle Operating Cost Savings

Travel Time Savings

Figure 10 illustrates project evaluation when generated traffic is considered. Congestion

Figure 10 Estimated Costs and Benefits, Considering Generated Traffic

Generated Travel Benefits

Figure 11 Road Widening Versus Transit Congestion Impacts

Alternative Transport Improvement Strategies

Considering generated traffic tends to increase the estimated value of improvements to

In 2020, the California Department of Transportation established specific requirements for

Framing the Congestion Question

References and Information Resources

Caltrans (2020), Transportation Analysis Framework: Induced Travel Analysis, California

CNU (2012), Highways to Boulevards Program, Congress for New Urbanism

Patrick DeCorla-Souza (2000), “Estimating Highway Mobility Benefits,” ITE Journal

Mark Delucchi (1996), “Total Cost of Motor-Vehicle Use,” Access

DfT (2007), Transport Analysis Guidance, UK Department For Transport (www.webtag.org.uk).

Anthony Downs (1992), Stuck in Traffic, Brookings Institution (www.brookings.edu); at

Miquel-Àngel Garcia-López, Ilias Pasidis, and Elisabet Viladecans-Marsal (2020), Congestion in

Roger Gorham (2009), Demystifying Induced Travel Demand, Sustainable Transportation

Induced Travel Calculator (https://blinktag.com/induced-travel-calculator), by the University of