The Cost of Slow Travel

Steven Polzin's picture

One of the most widely cited numbers in contemporary transportation media coverage and policy discussions is the cost of congestion estimates that Texas Transportation Institute (TTI) annually produces as part of the Urban Mobility Report series.   The 2009 version of that report (  shows an estimate of the cost of congestion of $87.4 Billion for the top 439 U.S. urban areas in 2007.  That estimate is composed of approximately $8.5 billion in wasted fuel costs, $14.57 in commercial vehicle congestion costs and $64.3 billion in personal time loss congestion costs.    

The same TTI report indicates the 439 urban areas have approximately 56 billion annual passenger miles of travel on public transit.  2007 National Transit Database (NTD) indicates transit vehicle speeds of approximately 13.4 mph while the 2009 National Household Travel Survey (NHTS) data indicates a door-to-door speed for transit travel of approximately 12.2 mph.  2009 NHTS data indicate roadway travel speeds averaging approximately 33 mph.  Using 12.2 mph for transit travel speed and assuming no differences in trip circuity (in reality the less dense transit network is likely to have slightly longer average travel distances for a given origin-destination pair), one can derive an estimate of the "extra" travel time required for using transit. 

Transit's slower average travel speeds result in approximately 3 billion hours annually of additional travel time.  If valued at the TTI time value of $15.47 per hour, this equates to approximately $44 billion annually in lost productivity due to travelers having or choosing to use transit.  Thus, the few percent of persons who use transit (approximately 2% of total person trips are on transit {5% of work trips} and approximately 1% of person miles of travel) incur 70% as much lost time relative to driving as is incurred by the total of auto travelers due to congestion, $44 billion for transit users versus $64 billion for driving in congestion. 

While one often hears about the "cost of congestion", there is virtually no one talking about the "cost of using slower modes of travel".  We hear a lot about the value of having a choice of modes but increasingly little about the value of having a choice of an uncongested or less congested travel option.  Is it fair to talk about the cost of using transit as $44 billion in lost productivity?  And what is the time cost of walking and biking versus alternative modes? 

Obviously these simple calculations do not capture all the interactions between modes or between land use and transportation that might occur with any changes in mode shares.  Also, the respective "costs" could arguable be recalculated with more accuracy using different time value assumptions or more geographically and temporally disaggregate data.  Nor does the data suggest I am recommending we should be discouraging the slower modes – in fact I am working hard to make alternative modes more attractive and competitive.  But it does hint that travelers are, in fact, rational beings and weigh factors that the planning community often neglects in much of the contemporary policy talk about modal performance and impacts. 

If time has value to travelers then it should have value regardless of the mode of travel.  While one could conjecture on the ability of travelers to use part of the transit travel time productively or divert the discussion to environmental or other impacts; transportation policy makers need to take the travel speed implications of various transportation policy initiatives into consideration if they are sincere in taking a holistic approach to evaluating transportation policy options and consequences.  Speed won't win in some of those policy tradeoffs but its importance to travelers and productivity should be appreciated.

One of the reasons the country and individuals have become more productive and the country has had growing gross domestic product over the past several decades is that we have been highly mobile and travel has gotten faster until recent years.  Part of the reason for faster travel has been the shifting from slow to faster modes and facilities.  There are lots of good reasons to enable and encourage use of alternative modes but analysis of the consequences should strive to be objective about the travel time and productivity consequences. 

Steven Polzin is the director of mobility policy research at the Center for Urban Transportation Research at the University of South Florida.



Accurate Measure?

Is travel speed really an accurate measure at all? Shouldn't time to destination really be the only factor? I understand your analysis wasn't intended to be all encompassing, but one glaring problem is that you are a using auto speed of 33 mph when that is overall average, not just for areas served by transit. Those areas will have slower roads, more stoplights, and more congestion.

I completely agree that an estimate of ALL costs should be included in cost/benefit calculations, but not just those bourne by the government. I'd like to see a cost comparison between a transit project and a road project that includes all externalities and long term maintenance, no matter how abstract they may seem. This includes public health costs (including stress), cost differences for the type of development each would encourage (utilities, residential streets, school busing, stormwater runoff, etc), personal auto maintenance and operation costs, accident costs (personal property, medical care, loss of life, civil court fines), cost difference of policing, etc.

Obviously this list could get out of hand, but in most cases transit wins the battle of 'stats that aren't counted'.

Mobility And Productivity

"One of the reasons the country and individuals have become more productive and the country has had growing gross domestic product over the past several decades is that we have been highly mobile and travel has gotten faster until recent years."

I don't think there is any evidence for this claim. US Productivity rose more rapidly during the 1950s and 1960s than in the subsequent decades, even though mobility and speed continued to rise as rapidly as ever during the subsequent decades.

If someone lives in a remote suburb and has a long commute on a high-speed freeway, why should that person be any more productive than someone who lives in an inner suburb or in the city and has a shorter commute? On the face of it, the claim is ludicrous.

Charles Siegel

Transit travel not as unproductive as driving

With cell phone networks, onboard wifi, laptops and smartphones, it's becoming easier and easier to be productive on transit. If you try to do the same thing while driving you'll kill someone. So things are not as bad for transit as your estimates suggest.

Travel speed as an excuse.

I appreciate any new idea to push automobiles and denigrate transit, and it is already known that transit takes longer.

But if all you are going to count is money, that tells us where you are coming from. If you are not counting anything else, we know you are likely ideologically married to autocentricity, which is boring. And it comes out in this argumentation.




I agree on the first part of your editorial which is that we should at least equally consider travel time regardless of mode in pricing societal "costs of congestion". To go a step further, I have always thought many of those things are mythical. Here is why: if I live and work and shop places and I already know how much time it will take to get here and there by any method, that is not a cost to society. You have internalized it so it's only a cost to you individually. Now, if there are traffic accidents, bus delays, a flat bike tire, those are unexpected travel costs to a large degree and so you can price them. But, I agree it's unfair to target every second of auto travel like it's a burden to society but if someone lese does the driving (bus, for example), you all of sudden recover time.

The last paragraph about economic growth and mobility I'm not so sure on. I don't disagree necessarily, but I don't see much here that really makes a case. I could be convinced, but I don't think that was the main point here.

A watched stoplight never changes

As a regular bike commuter/transit user and lifelong but now only occasional driver, and as a civic volunteer engaged in active transportation policy and planning--not a traffic engineer or transit planner--I see several holes in your argument.

Saying that time "has value regardless of the mode of travel" and then deliberately disregarding the multi-use nature of time spent in various modes almost feels like an oxymoron.

Yes, my time has value. The time I would spend driving to work gives me only driving (and frustration at stoplights, tailgaters and bad drivers).

If I'm on the bus reading or doing email, I'm part of the productivity you claim is due to speed.

If I'm riding my bike I'm getting exercise--time I don't have to spend at the gym later that night, taking time away from my family (or work, if you prefer) to do so.

It's quite common--particularly in congested urban areas--for bikes to be faster and more efficient than car travel for the short trips that make up the majority of urban trip counts. There's that productivity thing again, along with the fact that bike commuters are healthier and more productive workers.

You can't just factor in travel time either. What about time spent hunting for a parking place? My bus stops at the same place every time and the bike rack never moves--totally predictable elements in my transportation plan. Remembering where I parked my car at the end of the day? Having money or a permit for parking? That consumes time (and money I have to earn and then spend this way instead of on something more interesting).

I also would take issue with the idea that getting to work faster makes us more productive. You're implying/assuming that the travel time "freed up" by the car is spent at work, meaning we're working longer days.

If we're using that time to linger at the breakfast table and talk with our families then we're no more "productive" as measured by economists.

It may be true that we're all rushing out the door to speed to longer days at work but I hardly think that's a reason to praise the single-occupancy vehicle.

While you're at it, factor in the health care costs of our transportation system (as the American Public Health Association recently did, the productive days lost to automobile injuries and deaths (see this Canadian study for example, and I'm guessing the productivity numbers aren't so pretty. (Did you factor in unproductive time spent at the doctor's office for treatment of Type II diabetes and cardiovascular disease, both of which would be helped by increased use of active transportation?)

(I also think most economists give credit for our increased economic productivity to the advent of the personal computer and other technological advances--not to driving fast. You assert it here with no evidence.)

Your stated assumption that transit serves a longer and less-dense origin to destination network? I don't get that either; have any citations for this?

While there's a bit of chicken/egg debate, I think most people would agree that reliance on single-occupancy vehicles and the creation of homes farther and farther away from both employment centers and frequent transit service are correlated, at least in the West where I live.

It's less common to have sufficient density/frequency of transit service in those suburban developments, thus there's increased dependence on the car, thus the solo driving trips are much longer (and at higher average speeds) than the transit trips in downtown where stops are frequent and closer together.

If you're talking about people living two hours from downtown New York City and riding the train, that's because they chose to live that far from where they work. The speed of travel isn't to blame for the time consumed; it's the distance.

In my part of the world it's not train service that gets them to work, it's driving. Someone who lives an hour away from work and drives at freeway/highway speeds for the majority of that trip may be moving at a faster average speed, but he's not getting to work in less time than the guy riding the slower bus a couple of miles.


PS: My purely personal comparisons, both of which address time usage in my perception:
- bike vs. car hassle factor:
- car/bike/bus comparison:

Speed vs accessibility

You are equating speed with accessibility. To suggest that everyone-living where they do now-switch to transit would be absurd.

They need to move closer first!

Steven Polzin's picture

Speed and Productivity

The basis for productivity increases associated with mobility is not premised necessarily on persons converting the time savings to more work but that mobiity/speed allows people to travel farther and be more selective in employment and then be more producitve for themselves and their employeer. The "time is money" logic is decades old and born out by theoritical and behavioral data.

The issue of productive use of time by other modes may merit research. Some drivers use time productively with books on tape, cell phone use (surveys indicate ~8% of peak hour drivers are on their cell phones) and many folks eat in their vehicle and others use the time to think, relax, and plan. Some drivers cherish the drive time while others dispise it. The share of transit time that could be used productively is subject to debate as well. Transit users walk, wait, pay fare, transfer, watch for their stop, stand in the asile and hang on to the grab bars moving every time the vehicel stops - and some get a seat and can read or do something productive.

Speed, Density, and Productivity

"The basis for productivity increases associated with mobility is not premised necessarily on persons converting the time savings to more work but that mobiity/speed allows people to travel farther and be more selective in employment and then be more producitve for themselves and their employeer."

If densities go down at the same time speeds go up, people just have to travel longer distances to get to the same employment opportunities, and they do not have the opportunity to be more selective in employment. That is what has happened in our cities.

As the response below says, you have more employment opportunities if you have access to a variety of workplaces - not if you have to travel at high speeds across bleak, low- density landscapes to get anywhere.

Charles Siegel

On Productivity From Automotive News

"I don't think the car symbolizes freedom to Gen Y to the extent it did baby boomers or, to a lesser extent, Gen Xers," says Sheryl Connelly, who tracks cultural trends for Ford Motor Co.

William Draves, president of Lern, a consulting firm, says the digital age is reshaping the world much like the automobile reshaped American life early in the last century. The worker-bee mentality of a tight economy also is a factor, he says: "Time becomes really valuable to them. You can work on a train. You can't work in a car. And the difference is two to three hours a day or about 25 percent of one's productive time."

Read more:

PS on grammar: that should be "much as the automobile reshaped..."

Charles Siegel

Speed and Spaces

I agree with those who have commented that speed does not always mean more access. In fact, we could criticize automobility for squandering some *potential* accessibility (and thus productivity). Some of the potential accessibility benefits from speed are cancelled out by the fact that employment and other land uses get spread out as a result of relying on the car – from the need for giant parking lots, and the tendency toward fewer but bigger stores, schools, etc. In the theory, the most time-efficient scenario would be one where people use a fast travel mode between close places. Sounds more like a subway – up to the point where some trips are so short that nothing can beat walking or bicycling.

And, on the various Highway vs. Transit comparisons that come up now and then, I usually have this thought: the comparisons of current situations usually ignore how much one has been subsidized (highways, with money, land use decisions, and various other policies to enhance its convenience) and the other one has been straight-jacketed (transit, with lack of operating funds and incompatible development). It’s like saying: “Look, this plant with fertilizer and water is doing much better (on whatever measure) than that plant without fertilizer or water” and then trying to imply that the difference is actually the result of something inherent in the two plants.

The real question is: Which mode(s) do we wish to see performing well for our needs?

J. A. Love

A New Model is Needed

Peter Muller

Transit travel is inherently slow because it still uses the stagecoach model wherein riders have to travel to a stop and wait for the coach. The coach then has to stop whenever another rider wants to join or leave the trip. Finally the rider needs to travel from his/her destination stop to their ultimate destination. The only form of transit that improves on this model is personal rapid transit (PRT). PRT's off-line stations and small vehicles allow non-stop travel with almost no waiting. In addition extra stations can be added without slowing travel times. This, along with the ability to size stations according to demand, can allow stations to be placed at short intervals, thereby reducing station access travel needs and times.

The Morgantown PRT system has proven that this model works for over thirty years (a "White Elephant turned into a Transit Workhorse" according to the NY Times). A modern PRT-like system went into revenue service in Portugal earlier this year. The world's first true modern PRT system is scheduled for revenue service at Heathrow Airport this month, with another going into service in Abu Dhabi later this year. Learn more about PRT at

Engineer's Myopia

Gotta agree that singular attention to net travel time is absurd. Entirely too much energy pursuing productivity. Riding a bike, walking, riding a bus - there is real life-enriching quality associated with these activities, lost in cold analysis of traffic speed and travel time. Lighten up, please. Slowing down is not a "cost" to society - it is a measurable benefit to life quality.


Counting other things during the commute.

I had a much slower commute when I lived in Sacamenna for many years - my bike commute of 14 mi one way blew out all bad karma on the way in, and by the time I got home there was zero stress left from the day. And that is only the mental health aspect, not counting the low resting heart rate and the rest of it.



Slow Travel: Look Up

This essay does an excellent job of presenting the dilemma. One promising but over-looked piece is the transformation brewing at NASA: On-demand personal air travel at speeds 3 times faster than cars by aircraft that can truly be called "The Prius of the sky". Pocket airports can be built at half the cost of 1 mile of transit infrastructure.

A summary is given at:

and more can be found at:

measuring everything to mislead

Measurement is hot in corporate doublethink. The Exxon Valdez spill had a net positive effect on GDP. So that was good, right?

Wow. This is coming from a

Wow. This is coming from a director of mobility policy research? Aside from the deaths and injuries which present serious costs, from one's existence to all those extremely expensive hospital visits as already mentioned, what about the fact that even development geared 100% towards fast cars fails to provide a truly car-friendly environment? Isn't it telling that in the fastest growing regions where sprawl is always implemented that the traffic is so horrendous there, whether it's a DC burb or far north Columbus, you still get slow travel by unnecessarily exacerbating the distances between destinations and all of the single occupied vehicles and lack of alternatives to arterial roads. No one out there can just walk downstairs from their home and find any given destination in the first floor of their building, whether it be for shopping, a cup of coffee, or work.

Prepare for the AICP* Exam

Join the thousands of students who have utilized the Planetizen AICP* Exam Preparation Class to prepare for the American Planning Association's AICP* exam.
Starting at $245

Essential Readings in Urban Planning

Planning on taking the AICP* Exam? Register for Planetizen's AICP * Exam Preparation Course to save $25.
Grids and Guide Red book cover

Grids & Guides

A notebook for visual thinkers. Available in red and black.
City Plate table setting

New Arrival! City Plates

City downtown cores printed on gorgeous decorative collectible porcelain plates.