Car-sharing and the RAND corporation.


Today, Ezra Klein of the Washington Post highlighted a RAND Corporation report claiming that car-sharing cuts back very little on energy consumption.

One major flaw in the analysis: The conclusion is based on the principle that most car-share users are non-car-owners who already drive infrequently–fair enough. It then poses the hypothetical about 20% of the country switching to car-shares, and still claims that the energy savings are a drop in the bucket.

The flaw, of course, is that if 20% of the population were to switch to car-sharing, you’d see significant number of car owners using car-sharing services. But why would a car-owner use a car sharing service, other than perhaps to rent a vehicle type they don’t have in the garage (say, a SUV to haul lots of stuff)?

Well, they just might if they got rid of their car.

This is the key potential win for car-sharing services: it permits people to reduce the number of cars they own. Many people live in relatively transit-rich areas, and aren’t opposed to biking or walking–but still have cars and use them for most trips. Why? Ignoring cultural reasons, many either don’t trust the availability of the transit system, or have need to make trips (such as hauling lots of stuff) that can’t be made easily without a personal vehicle. And so they buy cars. And once they have a car in the garage, it is often more economical to go ahead and use it for everything.

The whole point of car-sharing services, from an environmental point of view, is that it enables people to sell their cars. Once you no longer have one parked in the garage, driving is expensive. But if you already own a car, you’re already paying most of the costs for it (depreciation being the biggest expense) regardless of whether you use it or not; at that point the marginal cost of driving is small.


15 responses to “Car-sharing and the RAND corporation.”

  1. By taking the AAA car cost data and processing it, the cost of a mid size car is $2000/yr to sure in your driveway plus about 25 cents per mile that you move it. That incremental cost is lower than almost any other form of transportation except very specific routes

  2. Normalizing variables another way, one could compute the cost of driving one’s car as about $10-$20 per hour. That means one must work additional compensating hours to pay for it. Further, time spent driving really is a self-externalized cost, for motorists never think about charging themselves for it.

    By this computation the real cost of motoring pushes $30-$50 per hour; the higher one’s wages the greater the cost!

    Look at motoring as an obsessive-compulsive disorder, much more difficult and dangerous than the productive labor one must engage in to be able to afford it. Gasoline was the 20th century’s overwhelming drug of choice, more generally addictive than smack, meth, coke, and the old standbys of nicotine and ethyl alcohol.

  3. If one tries to calculate the cost of lost time due to motor vehicle operation, it applies to rented cars as well. Of course, this is an understated reason why the wired generation seems to be less enthralled with cars than prior generations.

    And another good reason that wifi on transit is a valuable amenity to offer.

  4. Allan,

    Have you heard of a device called a bicycle? They cost about $0.01 per mile to operate. Using your numbers (which are actually very accurate, I have found), I save $5 each day that I ride my bike to work, and $2000 every year by sharing a car with my wife.

  5. Chris- great point. I love riding my bicycle. Some trips are slightly farther than I can ride. I just spent the weekend out of town playing at an ultimate frisbee tournament. Carpooling to the event was by far the cheapest option. However, I agree that for trips under 5 miles bicycling is way more fun and time-competitive.

  6. The $2000/year figure assumes that you buy a new car every 5 years or so, which is not true in our market, or with my generation. The depreciation/year numbers go down quite a bit but insurance stays about flat and maintenance goes up slowly :(.

  7. I don’t know. If you add up the costs of insuring a car, depreciation, and repairs that need to occur regardless of mileage (oil has to be changed at certain intervals, seals wear out even when it is parked, etc), I bet you can get pretty close to $2000 per year even with a used car.

  8. Scotty I feel like I’m a perfect example of what you’re talking about. Last year I finally felt like there were enough options with zipcar, transit, and my bike to sell my car and go without one all together. I drive soooo much less because the sunk cost isn’t there. If more people are like me, it’s likely to drive a huge reduction in VMT.

  9. We’re already seeing a drop in automobile use at the other end: young people delaying driver’s licensing and car ownership. The greatest environmental savings might be in the production of fewer cars, but at significant cost to the world’s manufacturing-based economy.

    The report also recommends advancing the prospects of autonomous vehicles. My main long-term concern about AV’s is the prospect of a major jump – upwards of 100% or more – in VMT, with most of that increase in vehicles operating without passengers or cargo. Families could eliminate the “need” for multiple cars if trips could be accomplished consecutively with the AV’s running empty between pick-up and drop-off points. Car-sharing AV’s could eliminate part, but not all of that increase.

    AV’s intrinsically have many environmental benefits which at least should partially mitigate those extra costs, including:
    1. the need for less parking
    2. the ability to locate parking well away from high value property
    3. the potential for reducing vehicle weight by the elimination of post-impact safety equipment and structural complexity
    4. the potential to reduce lane widths
    5. the potential to eliminate or at least minimize the need for many features of our road and street systems that are necessary with human drivers but would be superfluous with AV’s: signs, barriers, and even traffic signals, for example.
    6. the need to run transit vehicles in revenue service at significantly less than capacity.

  10. Car-shared autonomous vehicles will all but eliminate the need for private car ownership. (Well, except as an investment to generate revenue through a service like a “taxi” version of Getaround, or as a luxury item.) It will also eliminate public transit almost everywhere except a few places like New York City where density precludes putting everyone in a car.

    Smart phones will be ubiquitous well before autonomous vehicles are widespread. By then, transportation will become a simple matter of summoning a car on your phone (the car will come to you) and riding to your destination while you read or watch a movie or take a nap. I expect there will be various pricing options when you bid for a ride — for example, waiting for a less expensive vehicle, or getting a discount by agreeing to share the ride with others and taking a more circuitous route.

    Robot taxis summoned by GPS-enabled smart phones will change pretty near everything.

  11. I figure that it would cost me $5,000-$10,000 a year to own and operate a car sufficiently functional and reliable to be a daily driver needed to get me to work. This comports with the numbers from AAA.

    But I am retired, and my economy is much more viable without ownership of a motor vehicle. My annual expenses for local transport are about $300 for a TriMet “H” pass, $500 for operating a bicycle, $200 for the few times I really need a Zipcar. These sum to an order-of-magnitude less than cost of intensive motoring.

    There is an interesting thread going on BikePortland about the business-driven parking debacle on N.E. Multnomah. One commenter wonders why we give away parking spaces–which may well be more expensive than a car–but not the cars themselves:

    http://bikeportland.org/2012/05/16/pbot-unveils-plans-for-ne-multnomah-road-diet-71776#more-71776

  12. It made me feel better to seperate out the ‘costs of ownership’ from the ‘costs of travelling’ So often someone tries to bundle them together when they are really separate things and it is quite annoying to be told that it costs 55 cents per mile (AAA) or that it costs $5000-$10000 to own a car. These numbers are based on how much you drive. If only numbers were presented logically :(

  13. Douglas,

    Autonomous vehicles will help solve SOME of the problems associated with manually-operates SOVs:

    * Parking–vehicles can be parked out of the way
    * Improve congestion and safety–manually-driven cars should have performance improvements over those driven by humans.
    * Improve options for occasional car users–driverless taxis will eventually replace both traditional taxis and many short-term carsharing services (I expect there still to be a viable separate market for long-term car rental; but this too will be dominated by autonomous vehicles–in most cases, rental customers of the future will not be handed car keys, or even be provided with cars that have human controls).
    * Free up people from the chore of driving.

    But it won’t solve the following issues:

    * Energy use–a loaded bus or train remains far more efficient than a fleet of autonomous cars.
    * Cost–autonomous transit will be far cheaper to operate. Right now, car2go charges 35c/minute or thereabouts; I expect that short-term car rental prices to remain in that ballpark. With autonomous driving, the cost will drop as there will be no need to hire drivers to deliver cars to customers, and insurance will become simpler, but I expect that these services will still remain much more expensive than a bus or train ride (likewise assuming an autonomous vehicle).

    For more, see this article: http://portlandtransport.com/archives/2011/07/autonomous_vehi.html

  14. It’s also going to be difficult to overcome the gearhead, American independence mentality in this country. I honestly am not sure if I will see autonomous shared vehicles in my lifetime because of this. Our federal government can’t even properly fund transit/pedestrian infrastructure, or implement Positive Train Control safety measures, how are they expected to set up an environment where shared autonomous vehicles can operate? Then you have a constituency who will fight tooth and nail to give up their classic muscle car, or pickup truck. A surprising number of Americans actually plan for the destruction of our society, or the return of Jesus. How likely do you think it is that they will accept autonomous shared vehicles?

  15. Isn’t this just another type of gadgetbahnen, where the primary reason to push it is to kill off those pesky rapid transit lines (and, as lagniappe, avoid human interactions) and preserve the primacy of the American suburban lifestyle?

    It’s certainly not a scheme that’s well-suited for a densely packed city (all of those parking garages keep the density down; if the gadgetbahn was operational and all of them were converted into businesses, you’d have a considerable increase in the now two-way commuting traffic to get people there) and it’s not likely to change development patterns one bit in suburbia (people will want their robocars in their own driveways, and will want them right at hand at work or at the shopping mall, so the oceans of parking lots will not go away.)

    I’m sure that people could change their behavior to take advantage of the gadgetbahn, but it would be a lot cheaper and more efficient to change their behavior to take advantage of traditional mass transit, which is not being designed to take the existing auto-centric paradigm of American cities and turn it up to 11.

Leave a Reply to Allan Cancel reply

Your email address will not be published. Required fields are marked *