« $500 a barrel for oil? | Main | Blogging Death in LA »
The sources of fuel efficiency: a counterintuitive result
Matt Yglesias writes:
Via Andrew Sullivan, Eric dePlace notes that "You save more fuel switching from a 15 to 18 mpg car than switching from a 50 to 100 mpg car." And so you do. A 15 MPG car would require 1,000 gallons of gas to drive 15,000 miles while an 18MPG car could get it done in just 833 gallons. That saves 167 gallons of gasoline. By contrast, since a 50 MPG only uses 300 gallons to go 15,000 miles, upgrading to 100 MPG can't save that much gas -- the super-efficient car uses 150 gallons.
It is tricky, because the consumption basket and number of miles driven will not stay constant across alternatives, but this logic is worth keeping in mind nonetheless.
Posted by Tyler Cowen on December 23, 2007 at 06:37 PM in Data Source | Permalink
Comments
In some European countries fuel efficiency is measured in liters per 100 km. Say we did analogously and measured fuel efficiency in gallons per 100 miles. Then the 15, 18, 50, and 100 mpg cars require 6.67, 5.56, 2.00, and 1.00 gallons per 100 miles and the results make perfect sense.
In short, one needs to measure things in the "right" units.
Posted by: Isabel at Dec 23, 2007 6:52:51 PM
Agreed, miles-per-gallon is a backwards unit, and we really should be trying to minimize gallons-per-mile. After all, the goal is to use as little gas to go where you have to go, not to go as far as you can on a fixed ration of gasoline!
And for the policy wonks out there, U.S. CAFE fleet fuel efficiency requirements are actually computed 'correctly', by "by taking the reciprocal of the arithmetic mean of the reciprocals." ( http://yosemite.epa.gov/ee/epalib/incent2.nsf/02139de58cd4f6e18525648c00670434/06ef875972a435bd85256ab2007043bd!OpenDocument ) Otherwise, adding a few cars with very high milage could distort the fleet average MPG without reducing actual fuel use much.
Posted by: TomR at Dec 23, 2007 7:00:22 PM
It doesn't matter if we measure them in mpg or gpm, the result is "counterintuitive" (scary quotes required) because the percentual increase on efficiency can only give so much benefits. As your efficiency increases it's harder to get more savings, because you have much smaller expenses in the first place: 16.7% of 1000 is higher than 50% of 300. It's obvious if we track expenses instead of efficiency. Using the gpm measure the savings are the same: 1.11 (i.e. 6.67-5.56) is higher than 1.00 (i.e. 2.00-1.00) so the total saving from the first transition is still better than the second transition.
This kind of reasoning is useless anyway. What matters is the ROI of each choice. If the savings from both transitions offer good ROI why bother on how much are you saving (assuming there's no cost of opportunity)?
Posted by: Daniel Yokomizo at Dec 23, 2007 7:21:21 PM
It's ridiculous to debate in terms of percents here. You could also make the argument that going from 15 mpg to 100 mpg is more efficient than going from 15 mpg to 18 mpg. What's the benefit of this argument?
Posted by: Brad at Dec 23, 2007 7:24:35 PM
This is a pointless argument because all for vehicles are not compared side by side and give the illusion of counterintuitiveness.
15mpg car 1000gals/15000 miles
18mpg car 833gals/15000 miles
50mpg car 300gals/15000 miles
100mpgcar 150gals/15000 miles
of course there is a law of diminishing returns, but the counterintuitive argument would lend some less educated people to think that in order to help the environment and my wallet I should choose an 18mpg car over a 100 mpg car because it "saves more fuel" which is patently ridiculous. Assuming all four cars looked the same and performed at relatively the same speed, acceleration, braking and comfort, which car do you want?
Posted by: John at Dec 23, 2007 7:47:56 PM
Diminishing Marginal efficiency, Where is the counterintuitive part?
Posted by: Juan at Dec 23, 2007 7:48:00 PM
Some of you seem to be so educated that you miss the illusion that a less educated person might operate under. That is a good thing.
Increase my mileage by 3 mpg and I save 167 gallons.
Increase my mileage by 50 mpg and I save 150 gallons.
WHAT!?!
Well, you have to know the scale over which you make the change of course.
Does this have policy implications? Should we institute a minimum mpg standard instead of, or in addition to, the CAFE standard as calculated for fleet economy? Since relatively small improvements in mileage generate large returns at the bottom of the scale and relatively less at the top?
Posted by: Alan Coffey at Dec 23, 2007 8:27:38 PM
Why is this counterintuitive? Isn't it common sense that going from 1 million miles per gallon to two million would not save that much, because our usage would already be small. Just as if my cost of buying pepper were cut in half would not raise my real income very much.
Posted by: michael perelman at Dec 23, 2007 8:56:20 PM
I think Isabel's and Tomr's comments above and the Europeans are on the right track. Instead of MPG, cars should report GPM or gallons per 10,000 miles. This would make it clearer to the suburban family that its much better to trade in the old gas guzzler for a more efficient minivan than to replace a fairly efficient compact with a hyper efficient one.
More broadly, if you want to reduce oil use (and pollution), the focus should be on getting the gas guzzling cars and trucks off the road. The push by some politicians to have lower efficiency standards for "trucks" like hummers and SUVs, etc. than the the rest of car fleet can be seen to be especially foolish in this light.
Posted by: A student of economics at Dec 23, 2007 8:56:55 PM
Would people drive the same amount of miles or would they drive more miles in super efficient cars?
Posted by: Dano at Dec 23, 2007 10:17:23 PM
Would people drive the same amount of miles or would they drive more miles in super efficient cars?
Posted by: Dano at Dec 23, 2007 10:17:53 PM
As noted above this isn't so strange if you use litres/100km, as you should. But it is a useful reminder that fewer hummers is probably more important than more smart cars.
Moving to the US a few years ago, I was surprised how powerful the trucks and busses are: the UPS van has no trouble keeping up with the sedans. This struck me as strange, coming from a country where (compared to salaries) fuel is much more expensive, and so while some who can afford to drive cars with similar acceleration, trucks and busses seem to have smaller engines to save money. So I wonder how much room for improvement there is in commercial vehicles here?
Posted by: improbable at Dec 23, 2007 10:18:55 PM
More broadly, if you want to reduce oil use (and pollution), the focus should be on getting the gas guzzling cars and trucks off the road. The push by some politicians to have lower efficiency standards for "trucks" like hummers and SUVs, etc. than the the rest of car fleet can be seen to be especially foolish in this light.
This is true-- but at the same time is a caution against raising standards "too fast," if it results in people keeping older cars longer rather than buying new ones, as older cars get worse mileage just from wear. (Also, older cars have much, much worse emissions in other ways, because of the slow improvement in standards over the years.)
Another approach is to reduce the total miles of commute, which is why Mr. Yglesias has noted before that changing zoning to allow greater densities in city centers would help as well. DC is particularly bad in this regard.
Posted by: John at Dec 23, 2007 11:43:17 PM
Would people drive the same amount of miles or would they drive more miles in super efficient cars?
Generally, more miles, though the size of the effect is debated and it depends on your time scale. Miles driven is affected by where you live. Decisions about where to live, like durable goods purchases such as cars, are short term inelastic but more elastic long term.
The problem of driving more miles is one reason why a gas tax is more efficient than mandating higher fuel efficiency. Another problem are the various truck exemptions-- for certain jobs and situations the added power is needed and so some vehicles need to be sold as such, but once you make an exception everything under the sun will try to get in on it, as we've seen. A better solution is a tax; people who really need that power and towing capacity will pay for it, and people who don't will switch to more efficient cars.
Posted by: John Thacker at Dec 23, 2007 11:48:56 PM
"his is true-- but at the same time is a caution against raising standards "too fast," if it results in people keeping older cars longer rather than buying new ones"
Agreed. The more direct, and effective, policy is to simply tax the products with the negative externalities, e.g. pollutants, or near proxies, e.g. gasoline and other fossil fuels and let people buy whatever cars they want (as long as they pay for the externalities they create). Use the revenue to replace some of the taxes on activities with zero or positive externalities (labor, capital, research, etc) and its an obvious welfare gain.
Posted by: A student of economics at Dec 23, 2007 11:55:14 PM
Just use log miles per gallon.
Posted by: Ron Hardin at Dec 24, 2007 7:04:08 AM
I thought this was one partial explanation for why SUVs became so popular during the late 1990s and early 2000s.
Posted by: 8 at Dec 24, 2007 10:23:16 AM
What's the energy cost of making a new car?
Posted by: Nancy Lebovitz at Dec 24, 2007 11:01:44 AM
The "counterituintive result" is driven by a comparing MPGs (as a ratio), with savings in absolute gallons consumed. If you keep all units as percentages, logs or ratios, there is no puzzle at all:
Going from a 15 to 18 MPG car saves you 16.6% (3/18=0.166) on gas because GPM efficiency increases in 16.6% (167/1000 = 0.167)... And going from 50 to 100 mpg saves you 50% on gas because GPM is also reduced in half(150/300=.5)
Duh?
Posted by: Javier at Dec 25, 2007 1:36:54 AM
When people tell me that they can "afford" a new SUV, and what they really mean is that they can make the monthly payment on 6 year financing ... it saddens me.
If you can move that crowd off the dime, you might as well move them all the way to a Prius, but I'm not sure you can move them at all.
Posted by: odograph at Dec 25, 2007 7:52:23 PM
You would have to force people to change their preferences in order to change their lifestyles. I wouldn't live in my downtown "city center" if rent were completely free.
I agree about the diminishing marginal effeciency. What the efficiency remark says to me is that the biggest potential savings in fuel consumption is in the SUV and truck area. Moving someone from a V6 Accord (24 mpg) to Prius (45 mpg) nets an annual savings of 292 gallons. Meanwhile moving from a Durango (12 mpg) to a V6 Accord (24 mpg) saves 625 gallons.
So we see the rise in popularity of more fuel efficient vehicles. People are moving from Civics to Priuses, from Accords to Minis, from full-size SUV's to crossovers. I really don't see the need for government regulations, unless it includes a massive subsidy for the American auto manufacturers who are getting killed because they hedged all their bets on the longevity of SUV and truck sales.
Posted by: Jarick at Dec 26, 2007 12:05:43 PM
I think the MPG vs GPM question of superiority doesn't make sense.
Sometimes you know how much gas you have left and want to know how far you can go, sometimes you want to know how much gas it will take to take you a known distance. Each one will give you convenient numbers depending on the situation of the problem to be solved.
Posted by: anomdebus at Dec 26, 2007 3:04:23 PM
I think I talked someone into jumping from a Ford Expedition to a Prius this last week (14 mpg -> 46 mpg). It's not impossible.
Heck, I used to own things like Jeep Cherokees. I use a Prius for the same things, probably because like a lot of SUV owners I didn't really need it. I just kept waiting for that flood, or for the tree across the road.
(One time the road did flood, but the police waved me around, wouldn't let me ford it the Jeep anyway.)
Posted by: odograph at Dec 26, 2007 5:08:54 PM
Indeed, GPM is the "right" figure of merit precisely because it directly quantifies the amount of fuel used, whereas the inverse quantity MPG is counterintuitive owing to the diminishing returns discussed by others above.
Nevertheless, MPG is what American consumers are used to, which is why we chose miles-per-gallon-equivalent (MPGe) as a fuel-neutral measure of fuel economy for the Automotive X PRIZE (http://auto.xprize.org). For a graph that shows the diminishing returns visually, and for a longer discussion of why we chose 100 MPGe as the goal, see the FAQ "Why 100 MPGe, why not reward greater fuel economy" on page 27 of the Draft Guidelines (http://auto.xprize.org/files/downloads/auto/AXP_Draft_Competition_Guidelines_20070402.pdf).
js
Posted by: John Shore at Dec 27, 2007 9:37:44 PM
hi,I University majoring in the legal profession.After graduation,I 徵信 the work of the strong interest.Has worked in several徵信社.Has a wealth of experience. Now I immigrants France,Hope to continue to engage in the work of徵信 credit.
now, is to wake up every day to drink 咖啡, shopping. I hope that early awareness of Boles.
thanks,thank very much.
Posted by: Tonny at Mar 18, 2008 2:37:07 AM
