Reason Foundation

Reason Foundation

Surface Transportation Newsletter #116

VMT growth and the Millennials, Whither autonomous vehicles?, I-5 bridge collapse highlights flawed system

Robert Poole
June 12, 2013

In this issue:

VMT Growth and the Millennial Generation

For the last several years, popular media have featured articles about "peak car" and the recent decline of vehicle miles of travel (VMT) in developed countries. The anti-auto group PIRG (Public Interest Research Group) made a big splash last month, when their report ("A New Direction: Our Changing Relationship with Driving and the Implications for America's Future") got a favorable write-up in the New York Times (May 13th). PIRG's thesis is that the key to recent U.S. decreases in VMT and VMT per capita is changes in preferences and behavior among young people (ages 16-24)—the so-called Millennial generation. The story line is that Millennials are not very interested in getting a driver's license or living in the suburbs; they want to live downtown and walk, bike, or use transit. PIRG then does a lot of number crunching to produce VMT projections through 2040 that are drastically lower than anyone else's, and uses those results to argue for shifting a lot more transportation dollars from highways to transit, bikeways, etc.

In my column in the May issue of Public Works Financing, I questioned this thesis, drawing in part on analysis by commuting expert Alan Pisarski. He points out that the recession has hit young people the hardest, citing a Pew Center study in 2010 that found 37% of young respondents either out of work or underemployed. He also cited figures from the National Highway Transportation Survey that annual miles driven per person differ dramatically by employment status. For those 16-24, employed males averaged 12,000 miles in 2009, compared with about 6,000 miles for unemployed males. Pisarski also noted that in the last two decades, nearly all states have enacted "graduated licensing" schemes that restrict driving by those 16 to 18. Nevertheless, from 2001 to 2009, the fraction of those under 19 with drivers licenses was essentially flat.

In the May issue of The Atlantic, Derek Thompson assembled an array of data on Millennials, showing, for example, that the number of people ages 15 to 34 living with their parents soared from 0.5 million in 2008 to nearly 2 million in 2011. Fewer young people are getting married, having children, and buying homes—which has led to what Thompson describes as "stagnation" in household formation since 2007. As the economy recovers, all those trends are likely to reverse.

As for the alleged shift of Millennials and others from driving to transit, biking, and walking, that's very hard to find in the data. In a recent presentation at the U.S. DOT's Volpe Center, researchers David Pace and Don Pickrell compared the reductions in driving to increases in use of alternative modes. They find that increased transit use accounts for only about 1% of the decrease in auto travel, with bicycle and walk trips appearing to account for only a few percent more. Other possible factors (besides the recession) include the growth of online shopping and a nearly one percentage point increase in the mode share of telecommuting.

As for the much-touted preference of Millennials to live in the city center, the statistical evidence says otherwise. Demographer Wendell Cox reviewed the data last month in a piece for Comparing Census data for 2010 and 2000, Cox finds that people between 20 and 29 "were less inclined to live in more urban and walkable neighborhoods than their predecessors." In 2000, 19% of those in this age group lived in the core municipalities of major metro areas, compared with 13% in 2010. Likewise, in 2000 35% of them lived in suburbs, compared with 45% in 2010. (The balance lived outside major metro areas.)

So what can we make of all this? Clearly, VMT per capita cannot go on rising indefinitely. The major increases since World War II have come about via the entry of most of the female population into the workforce, along with rising prosperity that has enabled large fractions of lower-income people to afford cars. Pisarski and others predict that future growth in overall VMT will be driven primarily by population growth rather than increased per-capita driving. But the just-so story about Millennials losing interest in driving appears to be mostly an artifact of the recession's severe impact on younger people, not a fundamental change in their choices of where to live or how to travel.

» return to top

Whither Autonomous Vehicles?

As someone who has been a life-long science fiction fan, it feels strange to find myself being a skeptic on autonomous vehicles (AVs). While I certainly welcome continued advances in automotive technology, I think many AV enthusiasts have glossed over numerous problems and greatly exaggerated the potential benefits.

Many articles, even in serious transportation (e.g., Thinking Highways) and popular (The Economist) publications present a near-term future in which AVs have eliminated traffic congestion by tripling the number of vehicles that can operate on today's (unexpanded) highways and freeways. Car ownership will be replaced by car-sharing, eliminating all that time when costly vehicles sit in parking spaces, unoccupied. Some such articles portray a future in which the auto insurance industry has become obsolete (because AVs are so inherently safe) and in which drivers can snooze, play online games, or carry on business meetings while their AVs drive themselves. Drunk driving laws will become irrelevant, because it won't matter if you're drunk, since you won't be driving the vehicle. And so on.

A sobering reminder that things aren't that simple is "Proceed with Caution toward the Self-Driving Car," by Will Knight in the latest issue of MIT's Technology Review. Knight's main message is that "full autonomy remains a distant destination," due in part to human factors considerations. He cites cognitive scientist Don Norman's concerns that purely automated systems, without the capacity for driver override, can and have led to unsafe maneuvers (such as an adaptive cruise control that would automatically speed up when the car entered an off-ramp, since the lane ahead was now free of other vehicles). He also discusses MIT researcher Bryan Reimer's analogy with increasingly automated aircraft cockpits, which have led to reduced pilot performance when emergencies arise.

And Knight also points out that Google's much publicized AVs do not traverse terra incognita, figuring out how to navigate safely to the desired destination. Rather, then follow a route that has been painstakingly detail-mapped in advance—a process Google has only completed for a small fraction of the nation's roadways. Moreover, in a process so dependent on precise pre-mapping, sudden changes—such as snow obscuring signs, lane markers, etc. or the start of new construction—are likely to cause problems. The point is, just as in aircraft cockpits, there is likely to be a need for the human operator to take over control on very short notice under unexpected conditions—and there go visions of blind people driving or of playing computer games while the car does all the driving.

More sober AV advocates, such as Stanford's Bryant Walker Smith, acknowledge that we don't currently know what laws, licensing requirements, or product liability rules will be needed when AV technology has advanced to the stage of such vehicles actually being on the market. In a January article in the Wall Street Journal, Smith is quoted as follows:

"We don't actually have a fully self-guided car. We don't know how various aspects of that will be introduced. We don't know how they will ultimately perform or what 'safe' is. . . . It would be much easier to analyze the legal uncertainty if we had an actual product out there."

Then there is the question of privacy. Reason magazine columnist Greg Beato created quite a stir with his column in the June 2013 issue, "Google's Driverless Future." For one thing, he correctly pointed out that all the starry-eyed visions about no congestion and no accidents assume that all vehicles are AVs—assuming either 100% market penetration or the banning of non-AVs from some or all highways (as "Mothers Against Drunk Driving will morph into Mothers Against Driving, . . . determined to get America's deadliest assault weapon off the streets for good by advocating legislation that makes it illegal to operate traditional vehicles").

But Beato's even greater heresy was to suggest that Google may have an ulterior motive in promoting AVs. "In time, Google will know when you arrive at work each morning how many times a week you go to Taco Bell, how long you spend at the gym. . . . The driverless car is, in short, a data detective's dream. . . . In theory, Google will determine the route to your desired destination based on distance, available infrastructure, and current traffic conditions. But what if Google, which already filters cyberspace for you, begins choosing routes as a way of putting you in proximity to 'relevant content'?" A bit paranoid? Yes, but relevant now that we know the extent of federal government tracking of telephone and internet activity. I bring this to your attention because it's an aspect of AVs that I'd not heard before and that is certainly worth pondering.

Given this array of uncertainties, my guess is that we will not see truly autonomous vehicles on sale at auto dealers any time in the next couple of decades. Instead, we will continue to see the evolution of steps toward vehicle automation. Adaptive cruise control and lane-keeping systems, both of which assist drivers to operate their vehicles safely, are starting to spread from high-end vehicles to the rest of the fleet (just as anti-lock brakes and electronic stability control have). A very well-done film that dramatically illustrates the safety benefits of such technologies is "The Thinking Car." At 55 minutes in length, I put off watching it for several weeks, but once I sat down to preview it, I was hooked and watched the whole thing. Give it a try at:

» return to top

I-5 Bridge Collapse Highlights a Flawed System

This month I'd planned to write about innovative state DOT efforts on bridge repair and replacement. But the collapse of one span of the I-5 bridge across the Skagit River in Washington State calls for a different focus. I was away on a short vacation when it occurred, but as details came out in the days following the accident, I got more and more upset.

To begin at the micro level, Washington State DOT's attempt to shift the blame for the collapse to the trucking company is unconscionable. Mullen Trucking applied for and was given a permit to carry a load of up to 15 feet, 9 inches in height. That was despite the fact that the southbound vertical clearance just to the right of the outside lane was only 14 feet, 5 inches (with a maximum of 17 feet above the median). There were no signs ahead of the bridge warning of low clearance by lane, nor was there the kind of barrier one often sees prior to the entrance to a parking structure that prevents an over-height vehicle from entering. Apparently, WSDOT rules don't require warning signs unless the clearance is 14 feet, 4 inches or less. And a weasel-worded provision in the permit says that the state "does not guarantee height clearance." WSDOT spokesman Mike Allende was also quoted saying that "It's ultimately up to the trucking company to figure out whether it can get through. It's their responsibility to make sure the load they have can travel on that route."

Yet every engineer at WSDOT knows, or should have known, that the bridge in question is a through-truss bridge that is "fracture-critical"—i.e., the failure of any one structural element is likely to fatally weaken the entire structure. Hardly any bridges of that type have been built since the 1950s, since safer designs are widely available. Yet hundreds, if not thousands, of such bridges are still in service across the country, not all of them rated "structurally deficient" as they should be, given their inherently flawed design.

The broader question raised by this accident is why state DOTs aren't doing a better job of replacing structurally deficient bridges. Washington is hardly the worst offender on this score; its percentage of structurally deficient bridges is only 4.6%, compared with a national average of 11%. On the broader measure that includes functionally obsolete bridges, Washington is at 26.3%, compared with a national average of 24.9%. Far worse are states such as Pennsylvania (43.7%), Hawaii (44.6%), Massachusetts (52.9%), and Rhode Island (54.3%).

Rick Geddes, who heads the infrastructure program at Cornell University, rightly takes issue with WSDOT's attempt to shift blame to the trucking company in the Skagit River bridge collapse. "It is the responsibility of the bridge operator to ensure that oversized vehicles that may present a hazard stay off its bridge," he told me.  And in a May 30th article, he points out that were bridge operating companies responsible for such bridges, their contracts would "lay out clear performance standards, with associated penalties and rewards depending on measurable outcomes." (

Such contracts would require that state DOTs have the money to pay the contractors—and it is a shortage of funds that state DOTs are contending with nationwide. One answer, of course, would be to toll higher-traffic bridges to pay for their refurbishment or, in cases like the fracture-critical bridge on I-5, their prompt replacement.

» return to top

Surveys Explore Americans' Highway Fee Preferences

Three new surveys in recent weeks shed some light on what motorists and taxpayers think about paying for highways. In a nutshell: they over-estimate what they pay in fuel taxes and would rather pay tolls than additional fuel taxes to pay for improved highways.

A survey conducted by Ipsos for the American Road & Transportation Builders Association found that while Americans think safe and efficient transportation infrastructure is at least as important as other utility services (telephone, electricity, water, etc.), they think they are paying far more in highway fuel taxes than they actually pay. The average household pays $46/month in state+federal fuel taxes. But 24% think they pay about double that amount while 40% had no idea what they pay. Contrast this with the $124/month the average household pays for cable and Internet access, $160/month for electricity and gas service, and $161/month for cell phone service! This suggests to me that Americans would be more understanding of the cost of good highways if those highways were actually operated as utilities, charging motorists directly for how much highway services they use (presumably based on miles driven).

A second survey was HNTB's latest America THINKS survey, exploring attitudes toward tolling as a new or expanded funding source. It found that 71% would be willing to pay tolls to save time and bypass traffic congestion. Seventy percent said their state DOT should have the option to use tolling, and 80% supported public-private partnerships to deliver highway projects. The survey also explored attitudes toward priced managed lanes. Only 17% of those polled were aware of this concept, despite 38 such projects either in operation or under development in 10 states. But when the idea was explained to them, a whopping 74% said they would be likely to use such lanes if they had the opportunity to do so. Making that more specific, 68% said they would pay an average of $5 to save 15 minutes on their commute. And in contrast to fears of elected officials that such lanes would be seen as "Lexus Lanes," there was essentially no difference in acceptance of priced lanes between those earning less than $50,000 and those earning more than that.

In some respects even more interesting are the results of a survey of Wisconsin residents released by Marquette University. It polled them on various options for increased highway funding. Only 28% favored increasing the state fuel tax, and only 24% favored new highway bonds to be paid off with general tax revenues. But 53% preferred tolling for highway improvements. Majority support was found in every region of the state, except very liberal Madison, with support highest (59%) in Milwaukee.

My assessment of all this is that elected officials are out of touch with their constituents on transportation funding. They shy away from tolling (both North Carolina and Virginia legislators have recently voted against using toll finance to rebuild and modernize I-95 in those states), and they are worried that priced managed lanes will be seen as elitist.  We in the transportation research and policy community need to do a much better job of making the case to elected officials, to catch them up to what the public is already thinking.

» return to top

Bogus "Chinatown Bus" Study Exposed

Regular readers know I've been very impressed by the rapid growth of curbside intercity bus companies. The phenomenon started in the late 1990s as entrepreneurs began offering very inexpensive bus service from curbside locations in Chinatown in New York to destinations such as Philadelphia and Washington. Major bus companies responded to these upstarts by creating well-funded competitors such as Megabus (Coach USA) and BoltBus (Greyhound and Peter Pan). Overall, these companies have grown by leaps and bounds over the past decade, becoming the fastest-growing segment of intercity passenger travel.

But then came a 2011 study by the National Transportation Safety Board that claimed to find that curbside buses were unsafe. The headline number featured in dozens of news stories was that curbside bus companies were "seven times more likely" to be involved in a fatal accident than conventional bus operators. But that study has now been exposed as bogus by Jim Epstein, a producer at Reason.TV (a division of Reason Foundation).

NTSB stonewalled Epstein's request for the data on which its conclusion had been based. But after a fruitless six-month wait for the data via a Freedom of Information request, he managed to re-create the data from other sources (including a federal contractor whose accident database included a list of the 37 fatal crashes in the NTSB report). He received assistance in data analysis from Aaron Brown, a quantitative analyst at a hedge fund and also from a statistics professor at Wharton.

Epstein also obtained a list of the 71 bus companies NTSB classified as curbside and another 51 it classified as conventional. NTSB counted as "curbside" both Greyhound and Peter Pan, as well as several Trailways affiliates—all as conventional as they could be. Its conventional list included a large public transit provider, New Jersey Transit. Of the 37 fatal accidents allegedly occurring on curbside carriers, 24 were actually on conventional Greyhound. Overall, Epstein verified that 30 of the 37 accidents had actually occurred on buses operated by conventional carriers.

But it gets even worse. When the NTSB analysts calculated accident rates, instead of adding up all the fatal accidents in each category of bus provider and dividing by some kind of normalization factor such as bus miles, NTSB calculated a fatal accident rate for each company—large or small—and then averaged those rates! So not only were the buses mis-categorized, as noted above, but the rate calculations would have been meaningless even had the categories been correct. (You can read a lot more of the gory details in Epstein's account at

I have always had high respect for NTSB as a scientifically literate accident investigator. So this bizarre "study" must have some kind of explanation. Epstein speculates that political pressure led NTSB to come up with findings damning the curbside bus industry. He cites a long-running verbal campaign against curbside buses by Sen. Charles Schumer (D, NY), culminating in a letter that he and Rep Nydia Velazquez (D, NY) sent to NTSB demanding a safety study, alleging that it is "an industry that, in many cases, is operating outside the bounds of city, state, and federal transportation guidelines." Needless to say, when the agency's report was released in October 2011, Schumer immediately publicized the bogus "seven times" statistic. It's a sad day for transportation when even a well-regarded safety agency can be pushed into supporting a politician's agenda.

» return to top

Upcoming Conferences

Note: I don't have space to list all transportation conferences that might be of interest. Below are those that I or a Reason Foundation colleague are taking part in.

25th Annual ARTBA P3s in Transportation Conference, July 25-26, 2013, Grand Hyatt, Washington, DC (Robert Poole speaking). Details at:

IBTTA 81st Annual Meeting, Sept. 22-25, Vancouver Convention Center, Vancouver, BC (Robert Poole speaking). Details at:

» return to top

News Notes

P3 Project Performance Study. A joint study by Arup and Arizona State University compared the cost-overrun and on-time completion rates of 12 large highway public-private partnership (P3) projects with the results of studies of comparable projects carried out via design-bid-build (DBB) and design-build (DB) procurement methods. The results showed all but one of the P3 projects were completed on-time and 10 of the 12 were completed at or below budgeted levels. Both results were substantially better than for DBB and DB projects. A summary appears in the November 2012 issue of Public Works Financing, and the study itself may be obtained from

Maine Private Toll Road Delayed. A $2.1 billion private-sector project to link eastern Canada's Maritime Provinces to Montreal and Toronto via a 220-mile toll road across Maine has encountered a roadblock. In May, the legislature's transportation committee voted to repeal a 2012 measure under which the state would have paid for the feasibility study of the proposed highway.

Fresh Thinking on Oil and Gas. The May 2013 issue of The Atlantic includes a cover story by Charles C. Mann with the provocative title, "What If We Never Run Out of Oil?" It explains that the term "proved reserves" does not mean a fixed quantity but is instead a function of technology and cost. This principle—well-known to economists but often ignored by journalists and anti-oil folks—undermines the idea that the United States (or the world) has reached or will soon reach "peak oil." This article is well worth reading.

Federal MBUF Pilot Program Legislation. Rep. Earl Blumenauer (D, OR) is drafting legislation to implement a federal pilot program to test various forms of mileage based user fees. It would create a competitive grant program, analogous to the successful Value Pricing Pilot Program, to fund pilot projects that would test various ways to charge road users by miles driven. Blumenauer is seeking feedback from organizations like the Mileage Based User Fee Alliance on the draft bill prior to introducing the Road User Fee Pilot Project legislation

Lessons for America from Overseas HSR Projects. My Reason colleague Baruch Feigenbaum has authored a useful policy brief on the impacts of high-speed rail projects in Europe and Asia, focusing on such issues as ridership, environmental impact, economic development, mobility benefits, consumer choice, and competition with other modes. "High-Speed Rail in Europe and Asia: Lessons for the United States," is available on the Reason Foundation website. (

The Great Streetcar Scandal that Wasn't. Despite repeated debunkings, the myth persists that U.S. streetcar systems were dismantled by a conspiracy of auto and oil companies that bought up streetcar companies and replaced the rail vehicles with buses after World War II. In fact, ever since World War I, buses were less costly to operate, and far more flexible, than streetcars, and transit companies were replacing streetcars with buses all during the Depression years. Eric Jaffe does a good job of debunking the myth, drawing extensively on a first-rate article by Cliff Slater in Transportation Quarterly in 1997. You can read Jaffe's excellent overview in The Atlantic Cities, June 3, 2013. (

International Transport Forum Backs P3 Projects. At its annual meeting in Leipzig, Germany last month, transport ministers from 54 countries issued a declaration of support for public-private partnerships to help address the need for major investments to improve surface transportation infrastructure. The declaration called upon leaders to do a better job of explaining the pros and cons of P3 projects as well as placing a monetary value on roads and other transportation assets.

Leasing the Indiana Toll Road. Seven years after Indiana Gov. Mitch Daniels leased the Indiana Toll Road, what are the results? Reason Foundation's Director of Government Reform, Len Gilroy, provides a good overview, including how the lease proceeds were used, changes in the management and operation of the toll road, and political consequences for the Governor and the legislators who voted to approve the lease deal. (

Clarification re San Francisco Bay Area Express Lanes. In response to my April issue recap of priced managed lanes developments around the country, I received a long email from John Ristow of the (Santa Clara) Valley Transportation Authority. He pointed out that even though the MPO for the entire region, the Metropolitan Transportation Commission, has promulgated a regional express lanes plan, the lead on implementation is being taken by county transportation authorities, like VTA and the neighboring Alameda County Transportation Commission, which have introduced and are operating the first priced managed lanes in the region. I am happy to provide this clarification to readers.

» return to top

Quotable Quotes

"The cost of land poses a key dilemma for urban planners everywhere who want to concentrate jobs together so they can best be served by public transit. Such concentration raises the costs of land near the center; in fact, it would confer a monopoly advantage on landowners who owned such land and could exploit firms trying to locate there. Then firms will want to locate elsewhere to cut their land costs. Planned concentration of jobs in a few centers is not consistent with private ownership and control of land. Some type of collective control over land would be necessary to prevent monopolistic exploitation of land values. In theory this could be done with high land taxes and special zoning rules. But adopting those devices is politically difficult in a free enterprise economy. . . . A similar but less intensive dilemma concerns land near transit stops, where it would be most efficient to concentrate high-density housing and jobs. That also creates ownership monopolies over such land, unless it is specially controlled or taxed."
—Anthony Downs, Brookings Institution, "A Growth Strategy for the Greater Vancouver Region," 2007

"Indeed, the greatest promise of carbon capture is how it could transform the political debate. Unlike abandoning fossil energy, capturing carbon does not demand a radical alteration of national economies, global trade, or personal lifestyles. It does not threaten the ambitious development objectives of China and other poor countries. It enfranchises the very groups that have the most to lose from conventional climate policies—from powerful corporate interests to many of the world's poorest people. Ultimately, giving up fossil fuels simply may not be feasible, so we must explore alternatives. Carbon capture alone will not be enough to address climate change, but moving it toward the center of the energy-policy agenda will open up new opportunities for making real progress in the near future."
—Daniel Sarewitz and Roger Pielke, Jr., "Learning to Live with Fossil Fuels," The Atlantic, May 2013

"Since the drive to heighten the shipping clearance of the Bayonne Bridge comes from shippers who will benefit by the ability to use larger ships, why won't they pay 'tolls' to travel under the rebuilt bridge and help to pay for what benefits them? Truckers and other motorists get a slightly widened deck, but otherwise the main change they'll see is a longer, higher climb and more fuel consumed. Yet motorists are expected to pay for a benefit to maritime trade! That's a government agency in favor-trading mode, being inequitable as well as practicing bad economics. An eloquent argument for private investor ownership of a toll bridge like the Bayonne is that those in charge would be more likely to say to the container ship companies and the port, 'Our motorist customers are quite happy with the bridge as it is at 150 ft. If you want greater clearance for higher ships, we're happy to discuss that so long as you accept a responsibility of helping pay most of the cost.'"
—Peter Samuel, "PANYNJ Moves Ahead with Goethals Replacement and Other Big Staten Island Toll Bridge Projects,", May 15, 2013 (

"Many surveys have shown the public views 'user pays' as the most equitable way of paying for public goods that are individually consumed—more fair than paying for them out of the general fund or property taxes or myriad other less-equitable taxes. Essentially 'user pays' means treating roadways as a public utility, similar to water, electricity, or natural gas."
—Matthew Dorfman, "For Those About to RUC, We Salute You," Thinking Highways (North America), Vol. 8, No. 1, 2013

"When capacity (i.e. new lanes) is added to a roadway, it is immediately occupied by the existing latent demand in the rush hour(s). Or, the existing rush hour time period compresses and consumes the new [peak] roadway capacity. When new transit service is added in a corridor, roadway capacity that becomes available by that modal shift is again consumed by latent roadway demand. In other words, there is no capacity building strategy that can fully solve congestion, and even if there was, it is not affordable or sustainable. The only strategy that can solve for congestion is an operational strategy using the congestion pricing concept with deployment strategies like priced managed lanes. Not only do priced managed lanes provide a mobility option for automobile customers, they also provide a reliable transit corridor for buses at a much lower cost than traditional fixed-rail transit."
—Matthew Click, quoted in "HNTB Exec Makes Sharp Case for Toll Managed Lanes,", April 26, 2013 (

» return to top

Robert Poole is Searle Freedom Trust Transportation Fellow and Director of Transportation Policy

Print This