Highway advocates deny or minimize the science of induced travel

Induced travel is a well established scientific fact:  any increase in roadway capacity in a metropolitan area is likely to produce a proportional increase in vehicle miles traveled

Highway advocates like to pretend that more capacity improves mobility, but at best this is a short lived illusion.  More mobility generates more travel, sprawl and costs

In theory, highway planners could accurately model induced travel; but the fact is they ignore, deny or systematically under-estimate induced travel effects.  Models are wielded as proprietary and technocratic weapons to sell highway expansions.

Induced travel, or as its otherwise known, the fundamental law of road congestion, is a particularly inconvenient fact for highway boosters.  A growing body of evidence confirms what has been observed for decades:  adding more un-priced roadway capacity in urban settings simply generates more and longer trips, and does nothing to eliminate congestion.  Day by day, the popular media are starting to communicate this seemingly counter-intuitive fact to the public.

Highway boosters either simply ignore the entire concept of induced demand, or pretend that it doesn’t exist.  A new chapter in this effort to avoid this inconvenient fact comes from  Arizona State University Professor Steven Polzin, writing at Planetizen.

Polzin isn’t a complete induced travel denier; he’s more an induced travel apologist and minimizer.  It may be a real thing—or might have been in the past, he assures us—but it’s not a big deal and is now adequately being thought about by state highway departments and can safely be ignored.

Induced travel is scientific fact

Polzin derides induced demand as “a popular concept among urbanists” and argues that it’s given too much publicity in the media, by the likes of the New York Times.

But induced travel is not simply a “popular concept,” it’s a well researched scientific fact.  The best available evidence from a series of studies, shows that there’s essentially a unit elasticity of travel with respect to the provision of additional highway capacity.  A whole series of studies supports this estimate, some of which are shown here.

Duranton, Gilles, and Matthew A. Turner. 2011. “The Fundamental Law of Road Congestion: Evidence from US Cities.” American Economic Review, 101 (6): 2616-52.

Hymel, Kent, 2019. “If you build it, they will drive: Measuring induced demand for vehicle travel in urban areas,” Transport Policy, Elsevier, vol. 76(C), pages 57-66.

Hsu, Wen-Tai & Zhang, Hongliang, 2014. “The fundamental law of highway congestion revisited: Evidence from national expressways in Japan,” Journal of Urban Economics, Elsevier, vol. 81(C), pages 65-76.

Miquel-Àngel Garcia-López, Ilias Pasidis, Elisabet Viladecans-Marsal, Congestion in highways when tolls and railroads matter: evidence from European cities, Journal of Economic Geography, Volume 22, Issue 5, September 2022, Pages 931–960,

It’s odd that Polzin, a university professor, provides only a list of popular media articles (which he disbelieves) and provides not  not a single footnote or reference to a peer-reviewed academic study to dispute the notion of induced travel.

Purported mobility gains are an illusion

Sure there may be some induced travel, Polzin argues, but don’t overlook the benefits of greater mobility.  This misses the point that mobility (i.e. driving more and further) is evidence of induced travel, not a refutation.  And mobility tends to be short-lived and costly. Our friend and colleague, Todd Litman of the Victoria Transportation Policy Institute has a compelling rebuttal to Polzin on this point at Planetizen.

Polzin pleads with us to recognize the “mobility” benefits that come from increased highway capacity.  He misses two things here:  first, the key insight from the research on induced travel is that the mobility gains are at best a temporary illusion.  Somewhat faster moving traffic prompts more trip taking and longer trips, which quickly erodes any mobility gains.  And greater mobility simply prompts greater decentralization and sprawl, so even in places where traffic moves faster, everyone has to travel farther—and that comes at a real social, environmental and economic cost.

In effect, Polzin says that traffic growth is just due to population growth, and is inevitable, and good.  But he completely ignores the clear cross-sectional evidence from US metropolitan areas:  Metro areas with fewer lane miles of roads have shorter travel distances.  And far from being economically constrained, metro areas with less roadway capacity sprawl less, reducing public sector infrastructure costs, and creating a “green dividend” for their residents, who don’t have to drive as far.  The average resident of Portland drives about half as far every day as the average resident of Houston.  And, as we’ve documented at City Observatory, people who live in cities where people drive less are happier with their transportation systems.

Predict and provide = Prevaricate and pave

For decades, state highway departments have used their control over opaque and technocratic travel demand models to build a case for ever more highway capacity. Their “predict and provide” approach is the bureaucratic manifestation of induced travel.  Polzin never quite acknowledges this history, but instead suggests that we should simply trust highway planners to build new  traffic models that account for induced demand.

Much of the reporting on induced demand gives the impression that the transportation planning community is oblivious to this phenomenon or is comprised of road-building zealots. Newer activity-based transportation models are designed such that activity generation (trip generation) is sensitive to travel times. Consequently, improvements in travel speed will contribute to predictions of increased trip-making and travel distance. Even without the newest models, scenario testing and careful analysis of changes in demographics, mode choices, and flow volumes and patterns can give insight into the nature of demand on new facilities.

In theory, state highway departments could build models that accurately reflect induced travel.  But the simple fact is that they don’t.  To the contrary, a recent published article on the practice of state highway travel forecasting looked at this specific issue, and found just the opposite:  Induced travel effects are routinely ignored by state highway departments, and induced travel is generally introduced into highway environmental assessments only at the behest of public critics.  Those few state highway efforts that do consider induced demand, wildly understate likely effects.  Highway departments continue to produce models that exaggerate future travel demand growth even in the face of demonstrable capacity constraints, as Norm Marshall puts it “forecasting the impossible.”  And some, like Oregon, simply deny that induced travel is real, and prohibit their modelers from using scientifically based tools that estimate induced travel.

In a similar vein, Polzin solemnly intones that future transportation projects ought to be based on sound projections of future.

Roadway investments in new capacity should be based on up-to-date and sound demand estimates. They can’t just fulfill out-of-date plans or serve as ill-advised opportunities to create jobs or garner state and federal resources for local use. They should not use twentieth-century per capita travel growth rates or chamber of commerce-inspired population growth assumption

But there’s precious little evidence that state highway departments do anything of the sort.  They routinely plan for highway capacity expansions on roads where traffic is declining.  The Oregon Department of Transportation proposes expanding capacity at the Rose Quarter at a cost of $1.45 billion, even though traffic levels on that particular roadway have been declining for 25 years.  Cincinnati’s Brent Spence Bridge is slated for a massive $3.5 billion expansion, even though its traffic has been flat for more than a decade.  And other state highway departments routinely produce “hockey stick” traffic forecasts that are simply never realized.

The underlying problem that highway advocates fail to acknowledge is that road users will typically only use added highway capacity only if they don’t have to pay for it.    In the very limited instances in which drivers are asked to pay for even a fraction of the cost of providing increased road capacity, demand disappears.  The evidence from tolled roadways like Louisville’s I-65 bridge is that most people are unwilling to pay even a small fraction of the cost of freeway widening projects that would save them travel time.  That shows that the only reason people drive on expanded roadways is that someone else pays for them.  That’s pretty much the definition of induced travel.

Polzin’s piece is subtitled: “Induced demand is a popular concept among urbanists, but does its pervasiveness obscure the true costs of mobility?” This is a classic example of Betteridge’s law of headlines, the adage that states: “Any headline that ends in a question mark can be answered by the word no.”  Induces travel is real, and at this point, only highway advocates, and their apologists, like Polzin, are in any doubt about what this means.