You know you're a transport geek when you find yourself at a 9 PM debate about how governments should measure "urban mobility."
The opportunity arose one night at the Congress for the New Urbanism conference in Madison last week. Long after most urbanists had adjourned to the bars and restaurants, a small but sharp audience gathered to hear Tim Lomax of the Texas Transportation Institute (TTI) debate Joe Cortwright of CEOs for Cities. Tim was there to defend TTI's influential Urban Mobility Report (UMR), an annual compendium of statistics that are widely used to define how US cities think about mobility problems and to benchmark these cities against each other. Joe was there to attack TTI's methodology as biased against compact, sustainable cities.
The technical core of the argument is simple. TTI's Travel Time Index, one of their more quoted products, is a ratio of peak congested travel times by car against uncongested travel times by car. In other words, travel times are said to be "worse" only if they get much longer in peak commute hours than they are midday.
This ratio inevitably gives "better" scores to cities where normal uncongested travel times are pretty long — in other words, spread-out cities. Here's the CEOs' critique of how the TTI compares Charlotte and Chicago:
This index certainly looks hard to defend as any useful measure of travel time, even by car. Two cities that expose the average motorist to near-identical amounts of congestion delay are being scored entirely based on how far people have to drive there, where "further" means "better" on the travel time index score.
In his rebuttal, Lomax emphasized that Institute doesn't stress its Travel Time Index as much as it used to. He even offered a table showing the number of times his report mentioned the index in each successive year (plenty in the 90s, but way down in the late 00s.) We were to conclude that critiquing the Travel Time Index is a 20th century battle.
The real problem, of course, is TTI's title, "Urban Mobility Report," for a document that's really mostly about congestion. Only if you live in a very car-dependent city, or care only about car-dependent citizens, can you reduce mobility to congestion in that way. A more truthfully titled "Urban Congestion Report" would raise no objection.
Lomax argued that his insititute is interested in the other modes and wants to be able to talk about them more effectively, and I'm sure that's true. It's also true that nobody can be responsible for what journalists and editors choose to emphasize in reporting about their work. It may well be that journalists continue to fixate on the Travel Time Index more than the Institute itself wants them to.
But look at how the TTI's own website introduces its most recent study:
The 2010 Urban Mobility Report builds on previous Urban Mobility Reports with an improved methodology and expanded coverage of the nation's urban congestion problem and solutions. The links below provide information on long-term congestion trends, the most recent congestion comparisons and a description of many congestion improvement strategies. All of the statistics have been recalculated with the new method to provide a consistent picture of the congestion challenge. As with previous methodology improvements, readers, writers and analysts are cautioned against using congestion data from the 2009 Report. All of the measures, plus a few more, have been updated and included in this report.
Mobility is in the report title, but this entire leading paragraph is about congestion. If you dig into TTI's full press release about its 2010 report, you also find that it's all about congestion except for this curious paragraph:
The congestion reduction benefits of two significant solutions are discussed—public transportation and roadway operations. Without public transportation services, travelers would have suffered an additional 785 million hours of delay and consumed 640 million more gallons of fuel—a savings of $19 billion in congestion costs. Roadway operational treatments save travelers 320 million hours of delay and 265 million gallons of fuel for a congestion cost savings of $8 billion.
The journalistic spin that TTI itself recommends is that non-car modes matter only if they reduce congestion, and that congestion remains the primary measure of urban mobility.
What's more, TTI's suggestion that public transit directly reduces congestion is actually quite fraught, and many transit experts, including myself, steer away from it. Transit certainly creates alternatives to congestion for individuals, and the resulting benefit to individuals can be aggregated to describe society-wide improvements in both productive time and personal/family time. But those calculations are much more clear and direct than any "transit benefit to congestion" overall. That's because newly freed, high demand road space tends to induce new car trips.
Most transit projects are not trying to reduce congestion, or not all by themselves. If congestion reduction is your goal, you need a combination of transit and market-rate "decongestion" pricing for motorists. For most advocates of transit in the context of compact sustainable cities, the goal is not to reduce congestion but to give citizens options to liberate themselves from it.
Could a real "Urban Mobility Report" have value? Yes, but it would look totally different from TTI's, and it wouldn't be easy. Creating mobility measures that work across all the modes — cars, walking, cycling, transit etc — is fiendlishly difficult. I have some notions about how it might be done, but it would have to start with the right question. If you're going to talk about true urban mobility, then surely you have to ask questions like:
- How much of people's lives is lost to travel, where that travel has no positive value to them as personal time or recreation?
Or perhaps even more powerfully:
- What degree of freedom do people have to move about their city at will?
Any methodology that focuses on the performance of a single mode — whether congestion on freeways, continuity of cycle networks, or reliability of transit — is not going to lead us that way.
Given the resources and credibility that TTI has, I really hope they move in this direction. But it won't be easy. If the TTI report continues to be about congestion, that's another choice, but in that case their "Urban Mobility Report" will be a report of declining relevance (and increasingly offensive title) as this urban century unrolls.
Another question might be “how much of people’s time is spent (a) in transit [transport of all forms] + (b) earning the money to pay for that transport?”. If external costs are included, the answer would be even more complex. I’ve never read Ivan Illich, but remember someone summarising his argument that economically developed nations are spending more and more resources on transport in total.
Google is currently crowdsourcing the data from smartphones to get at real-time travel time and route frequency, but, get this, they have to distinguish the data coming from pedestrians and cyclists from those on the roads, so that they can use the data to update their real-time traffic data for secondary roads on Google Maps.
So…somewhere in the bowels of big-brother Google-land the capacity to get at the multi-modal mobility question is in development, if not already existing.
Sigh. According to the TTI, Chicago has much higher extra congestion per person than Charlotte, regardless of the travel time index. And average commute length is longer in Chicago than in Charlotte (and, more generally, in larger cities than in smaller cities). So CEOs for Cities has a lot more to explain about its methodology than “Chicago has lower total travel time.”
I’ve never met a commuter who was interested in travel speeds. What they’re interested in is total travel time, and sure travel speeds affect that.
But if you’re managing a city, and you’re interested in commuting, you should be focused on decreasing average commute times across the whole city.
You could try and do that by busting congestion (if that’s at all possible) or you could create an urban form with more local employment opportunities. It may not even need to be denser.
I loathe reporting on travel speed or congestion, and when I worked for a transport agency I always pushed the use of travel time over travel speed.
“Amount of congestion” aka travel speed is not a good measure of transport system efficiency, and “reducing congestion” aka making travel faster, is not a good measure of project benefits, without more information.
Urban consolidation in low density cities will almost certainly increase congestion, because it is unlikely that the desired extra transit use will fully make up for having less road space per person. It may still be worthwhile if it brings people closer to the places they want to be, to an extent that outweighs the slower travel.
An assessment of any city’s mobility will need to include movement by all modes, and also to account for multi-mode trips. Such an exercise will probably show that walking and cycling are pretty much congestion-free modes. What they lack is speed. And transit on its own rights-of-way is also pretty much congestion-free, but at a cost of longer walks and more transferring.
In a light-hearted vein, an easier way to get rid of congestion on regular roads is to lower the speed limit at each period of the day to match the actual speed, since congestion is simply a failure to achieve the posted limit.
And, as to the speed vs. travel-time argument, I find that speed is what people want, even if they use it to travel further, rather than put their time savings into a ‘bank,’ to be used for other activities other than being ‘en route.’
Gen Y seem to be coming to the realization that traveling further is not necessarily a ‘good’ and that achieving location-efficiency is a wise strategy. And they also are learning that time behind the wheel, even where the car is not going very fast, is downtime with regard to staying ‘connected.’ Driving is a big sink-hole. The statistics on driving distances is showing we have achieved ‘peak car.’