Does Transit Save Energy?

This is one of those questions that seems like a no-brainer -- a bunch of people are sharing a ride, so they must be saving energy.  When asked this question, we all think of a full bus or train of people vs. the number of cars that would have carried the same people.

The key issue turns out to be occupancy -- how full is the train or bus.   And it turns out that occupancy is probably lower than most people think.  That is because everyone rides on buses or trains as they commute -- they are going in the direction of most people's travel at the time of day they travel, so the transit is totally full.  But no one thinks about those trains having to go back the other direction, usually mostly empty.   As a result, we get to this fact, from the National Transit Database as synthesized by Randal O'Toole.

2014 Energy Use per Passenger Mile

  • Transit:  3141 BTU
  • Driving:  3144 BTU

Valley Metro Rail here in Phoenix does better, at a reported 1885 BTU per passenger mile.   As reported many times here on this site, the cost of building this rail line, now well over one and a half billion dollars, would easily have bought every round trip rider a new Prius, with a lot of money left over.   This would have saved more energy as well.  Buses in Phoenix are averaging just over 6000 BTU per passenger mile.



  1. Bram:

    Nobody ever thinks about the massive amount energy it takes to build a railroad and the trains themselves.

  2. Jim Collins:

    "now well over one and a half billion dollars, would easily have bought every round trip rider a new Prius, with a lot of money left over. "

    But then that $1.5 billion wouldn't have been available for all of those union jobs building, operating and maintaining that system. There also wouldn't have been any money for paying back political supporters. I can go on and on.

  3. Onlooker from Troy:

    Crap! You introduced me to yet another good blog that I can't keep up with! :-)

  4. DirtyJobsGuy:

    Transit will come out worse than driving if you consider most of driving is portal to portal, while transit often requires driving to a station then walking/taxis to your final destination. This is especially in areas like Boston or DC where commuters drive and park at a metro station. It astonishes me that while we have a well developed system allowing users to go point to point easily on schedules they choose transit backers want to impose a much more inflexible system

  5. Daublin:

    Enjoy. The Antiplanner is really fun. He was involved in writing transit plans in the past and then for lobbying for them, and he became disillusioned by what he saw.

  6. xtmar:

    Walking seems like it would have very low energy per mile, as the infrastructure requirements are relatively low, and people are very efficient walkers.

    I think the argument for transit in the very densest cities (NYC, London, etc) is that even if it's more costly or energy intensive ( which is questionable in such dense areas) it is the only practicable way to bring people into the center city in reasonable numbers, because there simply isn't enough parking or lane miles of highway for everyone who works in Midtown to drive. On the other hand, for relatively less dense places like Houston or Denver, is basically just to attract yuppies

  7. joe:

    HOV lanes are also big boondoogles

    Consider the follow - three traffic lanes carry approximately 2x the number of cars vs 2 lanes
    4 lanes will carry approx 50-60% more traffic than 3 lanes.
    All the HOV lanes do is impede the flow of traffic for the benefit of 2% of the commuters than travel with two passengers. The overall result is more gas burned and more pollution, and considerable costs to manage/maintain / the legal enforcement costs of ensuring that every car has at least two passengers.

    Secondly, due to demographics, work locations, residential locations, etc, nearly half the users of the hov lanes are not commuters to work.

    construction companies love them because of the all the revenue for them.

  8. ErikTheRed:

    You don't mention whether these BTU numbers are inclusive of the hot air from politicians and greenies breathlessly hyping these projects.

  9. marque2:

    Sure, but they would be auto union jobs.

  10. jon49:

    Occasional in the past I would use the buses to get around. It would be a huge bus with just a few people on it, in the middle of the day. It would be nice to get rid of public transit and then get rid of all the crazy regulations around private companies, then maybe someone could have some sane "mass transit".

    Oh, and it looks like CA is trying to kill the driverless cars. hhhrrrmm.

  11. markm:

    It's not just the trains or buses going back empty, it's them running nearly empty during the off hours. There are several reasons it keeps going all those hours when there are few customers:

    1. Trips other than commuting: What the mass transit promoters want is to persuade people to give up their cars. That won't happen if the buses only run during weekday rush hour. People also have to get to the doctor. They have to go shopping. They want to go out at night - and that's expensive these days enough without also having to pay for a taxi home. If the mass transit system does not meet these needs, most people will keep their cars, and eat the expense of insurance, parking, and maintenance. Once most of the car expenses are sunk costs, why pay more to take a slow, crowded ride from a bus stop blocks from your house to another bus stop blocks from your workplace? So to try to meet all those varying schedules, mass transit ideally will run 24x7, or at worst will start at the earliest point of morning rush hour and run late enough to take a drunk staggering out of the bar at closing time, and also run on weekends.

    2. Varying shifts: Not everyone works 9 to 5 (or 7 to 3, which is the common daytime factory shift). If commuters are going to pick up coffee on their way to work, someone else has to come to work even earlier to open the shop and load the coffee brewer. Others have to work in the evening, so people can shop and pick up dinner after work. So the buses must start at 5 am and keep going until 9 or later, or they aren't even serving the low-paid workers who need them most.

    3. Overtime: So maybe you decide to implement mass transit only for the 9-5 workers in those skyscrapers downtown. (This also greatly simplifies the system routes, since one end of every trip is in a small area downtown; instead of a grid, it can be a star.) That's fine if no one ever unexpectedly has to stay late - so it would probably be fine for unionized government workers. I suspect that if a typical government office received word at 5 minutes to quitting time that a volcano was going to erupt in a populated area that night, the warning wouldn't be passed on until next morning. But most salaried people in businesses these days must plan to be able to handle emergencies at work, staying hours later if needed. They don't know when an emergency might arise, so if they can't catch the bus home at 9 or 10 PM, or possibly even later, they have to drive _every_ _day_. Even hourly factory workers may arrive at work to learn that because of a rush order or an unexpected interruption in the schedule, everyone is working overtime.

    4. Transit worker shifts. You can't schedule unionized bus drivers to work only 2 hours in the morning and 2 hours at night, anyhow. You have to give them mostly 8 hour shifts, so they'll be driving at the deadest part of the day.

    5. Capital expenses: A bus system _could_ save a lot of fuel by switching from huge buses to small ones in the off hours. But then you have to buy twice as many vehicles - a perpetually money-starved city (because the damfool politicians spend every dollar they think they'll have their hands on, often before they actually have it) won't be able to afford that. (This has come up and been rejected repeatedly in my town.) Trains might be able to drop off the extra cars in a siding, but my impression is that modern subway and light rail systems have made detaching and reconnecting cars considerably more complex than it is for freight cars. So rush-hour capacity vehicles are running all day and on to midnight.

  12. markm:

    If the advocates for mass transit were actually looking for more efficient ways to get people to work, they'd be pushing to implement self-driving cars as soon as possible. That would enable automated taxis, at a fraction of the cost of taxis with human drivers, and less than the real cost of mass transit. 100% automated lanes would clear up a lot of congestion, and taxis or individually-owned automated cars would not need to find parking downtown. If the transit-pushers want to call those automated taxis "transit pods", that's fine with me, as long as they don't make it a government monopoly.

    But that doesn't give the progressive-fascists what they really want: control over the people. With a train, and even with buses, they in effect tell people, "You will live _here_ and work _there_, on the schedule we decree," because their goal is to make it impossible to get to work otherwise. (As that proto-hippie Henry David Thoreau put it, "The iron horse is in the saddle and riding mankind.") Automated transit pods will go where people call for them, when they're called. The commoners will have the same freedom to travel as a limousine liberal - and that'll really chafe those who want to be our superiors.

  13. Noumenon72:

    Is there some kind of traffic flow equation leading you to say 3 lanes carries 100% more traffic than 2 lanes, or are you just having math problems? Normally 3x is 50% more than 2x, and 4x is 33% more than 3x.

  14. Baelzar:

    We've got the Rail Runner; 28+ million a year to maintain, 2+ million a year income. 780+ million balloon payment up soon. Sold to our state with lies and obfuscation, but mostly with the help of the idiot leftists who live here.

    Of course now they explain, "well of course no transit solution ever makes money." A huge anchor around the neck of one of the poorest states. The leftists love it.

  15. ano333:

    How does that compare to building the roads and cars?

  16. joe:

    Noumenon72 - Is there some kind of traffic flow equation leading you to say 3 lanes carries 100% more traffic than 2 lanes, or are you just having math problems? Normally 3x is 50% more than 2x, and 4x is 33% more than 3x.

    you are obviously not a engineer - its a well known concept with air flow and liquid flow systems. an 8" pipe carries approx 30%-40% more fluids than a 6" pipe, There is a certain amount of friction & turbulance that has to be overcome. As the pipe gets larger, the friction is reduced. Same with car traffic. A single lane only travels as fast as the slowest car, (slow pokes, cars turning etc.) As more lanes get added, the slower cars dont impede the flow of the faster cars in the other lanes, ie less friction.

  17. slocum:

    "Walking seems like it would have very low energy per mile"

    Actually, it doesn't (once you consider the amount of energy required to produce the extra food consumed):

    But it depends on what you eat (and biking is much efficient).

  18. Matthew Slyfield:

    There is another factor that needs to be considered. The ratio of energy needed to move the passengers / the energy needed to move the vehicle itself. the larger the vehicle gets, no matter how much you increase passenger capacity the ratio gets smaller.

    For a passenger train, that ratio approaches zero.

  19. xtmar:

    An 8" pipe also has almost twice the cross sectional area of a 6" pipe, (16pi vs 9pi sq in) because of the linear/square/cube issue, so it's not surprising that it would carry 40% more fluid.

  20. xtmar:

    Very interesting! Thanks!

  21. MJ:

    In the case of Phoenix, it's important to look at what happened to bus transit use to evaluate the impact of light rail. Phoenix has one of the more heavily-used LRT lines in the country, but that's largely because it has replaced heavily-used bus routes in one of the most heavily-used bus systems in the country.

    For example, I took a look at the average boardings per revenue hour (revenue hours measure the time the buses are actually in service and are a fairly common measure of productivity or service effectiveness for transit systems) for Valley Metro/City of Phoenix both before and after the opening of LRT. Warren has noted in previous blog posts that the decline in bus use seems to correlate with the opening of LRT service. Looking at the year 2005, an arbitrarily chosen point a few years before the opening of Phoenix light rail, the Valley Metro bus system averaged about 115 boardings per revenue hour systemwide -- this is a very heavily-used bus system by US standards. In 2013, the most recent year for which data were available, that figure dropped to about 87 boardings per revenue hour. That's a fairly precipitous drop (almost 25%).

    I haven't tried to calculate the effect on average energy efficiency of Phoenix's buses, but suspect it's largely proportional to ridership. This is a common outcome with the opening of LRT systems -- there is an coincidental decline in bus productivity, as light rail replaces some of the region's more productive bus routes and relegates them to feeder services for light rail with correspondingly lower demand.

  22. mx:

    Yes, because cities like New York City or Boston where transit is popular are full of bureaucrats determining where you can live and work and at what time. Mass transit is a network and some connections at some times are more convenient than others. Similar, roads are also a network and some connections at some times are more convenient than others. The decisions about what buses and trains run when are no different from the decisions about how big the roads should be and how much parking there should be.

  23. Noumenon72:

    Your logic is solid, but your example of the pipe isn't great, since an 8" pipe has a cross-section 33% larger than a 6" pipe, so what would you expect? But I believe you.

  24. TruthisaPeskyThing:

    I am not sure that fluid dynamics are the best analogy for traffic capacity of various # of lanes. I believe in electricity, increases the wire's diameter from 6 to 8 centimeters increases the current by 78%. But then the cross-sectional area of the wire by 78%.
    Nevertheless, Joe, is correct that traffic flow increases proportionally more than the number of lanes. Congestion is caused by slow drivers, distracted drivers, turning drivers, nervous drivers . . . and increasing the number of lanes increase the opportunity to move around them.

  25. fd:

    I've wasted more life that I care to contemplate idling in a car at a crossing while waiting for empty trains to pass. Has anyone ever quantified the energy wasted by vehicular traffic that has to stop, idle, and start again every time another empty commuter train goes by?

  26. Peejay70:

    It's not "control over people" they want. It's to kill the automobile. They have an almost religious antipathy toward cars. Control over people is a means to an end. They have other objectives as well, of course, as bans on sugary drinks and trans fats attest.
    Here in DC they support crazy ideas like the DC street car - despite the fact it will take nobody off the roads - because they think it's a first step toward killing the automobile. They are nuts.

  27. Johnathan Swift Jr.:

    The ultimate monument to this type of stupidity will be Jerry Brown's Crazy Train to Nowhere! It will break every record in terms of spending and lose more money per mile that any system yet built by the malignant idiots known as urban planners and mass transportation advocates. The chances are that it will be started and then when the cost overruns and the ginormous cost and complexity and risk of tunneling through the San Gabriel Mountains becomes too much, simply abandoned, truly a train to nowhere, a true monument to the repellent egotistical narcissist known as Jerry Brown.

  28. Johnathan Swift Jr.:

    That's Bloviating Turkey Units.

  29. SMM:

    There is an equation, but it's complicated by all the necessary
    adjustments. Vp = V / [(PHF)(N)(f hv)(f p)] . Used in conjunction with a level of service curve. N is the number of lanes. So the demand volume per lane Vp is inversely proportional to the number of lanes (N). The level of service of a lane of traffic is inversely proportional to the demand volume. Level of service refers to the relationship between density of cars and the flow capacity of the roadway: the more cars trying to go down the road at the same time, the slower the flow of traffic - you see this in practice everyday during rush hour. If you try to put 3000 cars per hour down a one lane road that can carry 2400 cars per hour, you get a logjam, and the flow is lower because the speed is lower. Put 3000 cars per hour down a two lane road where each lane has capacity for 2400 cars per hour, and traffic flows much better. So, in short, adding an additional lane to a congested situation (such as opening an HOV lane up to general traffic) increases the capacity of the roadway more than proportional to it's lane space, by reducing the density of cars in all the other lanes on the roadway, which increases the speed and flow in all the other lanes in addition to the new capacity.

  30. SMM:

    Also, converting HOV lanes would probably decrease emissions as well, since all those cars that are now able to move will be getting much better fuel economy then if they were sitting still watching a handful of cars fly by in the HOV lane.

  31. SD Sun Devil:

    I can't quote any particular study, but I've heard that the most efficient way to spend transportation dollars is simply to add more lanes to roads. Mass transit suffers from massive inefficiencies outside of peak times. As stated, even during peak times, most passengers are going one way.