How MPG Misleads

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.

One moral of the story is that the MPG statistic is probably misleading a lot of people who aren't quantitatively sophisticated. In policy terms, meanwhile, the upshot is simply that it makes more sense to focus on raising the efficiency of the least-efficient vehicles than creating new super-cars. Of course, the genius of pricing carbon through a tax or through auctioned emissions permits is, once again, that is spares people the burden of trying to do all the math in our heads and just lets price signals automatically find the most economical way of reaching the targets.

Comments

> One moral of the story is that the MPG
> N statistic is probably misleading a lot
> of people who aren't quantitatively sophisticated.
> In policy terms, meanwhile, the upshot is simply
> that it makes more sense to focus on raising the
> efficiency of the least-efficient vehicles than
> creating new super-cars.

As long as you ignore the ownership cycle and fleet replacement issues, sure. National fuel economy kept rising in the early 1990s even as the horsepower/SUV craze hit because the more efficient late 1980s cars were still working their way through the fleet. Ban SUVs in 2008, build supercars in 2009, and by 2015 national fuel consumption will really start to drop.

Cranky

Better yet, put in an "inspection" law for SUVs similar to the one Japan has that is designed to force cars into the recycling center in 5 years.

Posted by Cranky Observer | December 23, 2007 2:00 PM

The problem is that you can't look at these variables one at a time-- the various parameters tend to clump and correlate in unpleasant ways. So, e.g., a multi-vehicle household in exurbia is going to use a lot of gasoline even if the vehicles are more efficient.

Posted by MattF | December 23, 2007 2:02 PM

There is a massive difference between a carbon tax and an auctioned emissions system. Even within the auctioned emissions camp, you run the spectrum from cap-and-grandfather (which is total smoke and mirrors and will do nothing) to things which probably would have an impact but can't do what needs to be done. Proponents of the first type are industry hacks, and proponents of the second type of trade system insist that it's just a first step but they can't go past that because the resistance from industry would be too stiff...but that's never going to change and there would never be a second step.

Posted by Ed Marshall | December 23, 2007 2:11 PM

One thing the car companies have been overwhelmingly successful in doing is creating the FALSE impression that raising fuel efficiency (yes, even dramatically) is in someways difficult. It isn't.

Cars have been been making leaps and bounds in efficiency for years. Unfortunately, because CAFE standards remained unchanged, the car companies had no incentive to plow those efficiencies back into improved gas mileage. Instead, they built bigger, faster, high horsepower vehicles that accelerate well - and this is how they marketed the vehicles.

So, a dramatic elevation in mandated gas mileage would not be hard to implement. Yes, it might mean the elimination of armored personnel carrier sized SUVs, it might result in cars that can no longer go 200 MPH, or that can accelerate from 0-60 in 4.5 seconds, or from 40 to 60 in 1.2 seconds...but, by the same token, these seem like rather luxury abilities, and, after all, it is the planet we are talking about.

Posted by McKingford | December 23, 2007 2:45 PM

I should also point out the fallacy in the contrast here: we aren't talking about getting people in Honda Civics to drive Honda Civics that are twice as efficient, but getting people out of Lincoln Navigators and into Honda Civics (or, at least, Accords).

Posted by McKingford | December 23, 2007 2:48 PM

I wouldn't want to see SUVs banned, just taxed or otherwise made expensive enough that people who don't need them buy something else instead (hey, remember station wagons?) SUVs weren't invented in the 1990s, it's just that that's when they began to become so popular. But before that they were seen for what they should be seen as: a specialty vehicle, useful for forest rangers, lifeguards, geologists, etc.

Posted by Adam Villani | December 23, 2007 2:59 PM

"One thing the car companies have been overwhelmingly successful in doing is creating the FALSE impression that raising fuel efficiency (yes, even dramatically) is in someways difficult. It isn't."

Total complete b.s.

The big 3 know how to build fuel efficient cars - they've been doing it for years in Europe. (e.g. the top selling city car in the UK is the Ford KA). The problem they have isn't building them, it's selling them in the U.S. where fuel costs are so much lower that people just don't care that much about fuel efficiency.

Posted by kafka | December 23, 2007 3:10 PM

Any talk of conservation is at best a half-talk unless mass immigration is included in the discussion.

The US Census says legal and illegal immigration is responsible for 90% of our population growth and 85% of our energy demand. We have gone from 225 million in 1973 to 300 million today to US Census projections of 363 million by 2030 and 420 million by 2050. Until the 60s, when population growth and more, and more gas-hungry cars and other energy consumption caught up with us, America was a large petroleum exporting country.

All the major energy savings of the 70s were cancelled out by 1985 by all the Juans and Pedros pouring in.

But other regions have it far worse for both energy conservation and resources given their high breeding rates. Africa went from 100 million people in 1900 to 850 million today, with a net decline in standard of living.

Many environmentalist groups decided to take immigration "off the table" of any discussion of habitat loss, defeat of energy conservation efforts in the 70s as being "racist and non-PC". Carbon taxes? Boosting Juan and Pedros F-150 gas mileage? Less SUVs? Doesn't matter if we see world population explode and a good portion of that run across our BOrders - all those conservation measures will be negated by even 3 million illegals getting legalized...

It is also worth mentioning that the same people that want to end filthy fossil power to save the planet, besides being willfully blind to population growth - completely reject nuclear power.

It is also worth mentioning that the technologically ignorant think SUVs are a huge part of our energy use. They are not. The US uses 108 Quadrillion BTUs of energy. 16 in private transportation. 7 involving use of SUVs and gas-guzzler high performance cars, RVs, recreational boating. Improve gas milage 100% (not really doable for a reasonable price) and all you achieve is 105 Quads while all the Juans, Abdullahs, Marias coming in and their kids will add a demand for another 25-30 Quads by 2050.

Posted by Chris Ford | December 23, 2007 3:11 PM

I've been annoyed for a while that my car can give instantaneous MPG and km/L but has no option for other units. If you care about efficiency, the logical units are something like the reciprocal: either volume / distance, or, even better, volume / time. If you do that, then your brain can have a decent chance of averaging out performance and understanding what difference it makes to drive one way or another. MPG is logical for the purposes of making sure you make it to the next gas station, which was probably more important at some point, but terrible for guessing the efficiency of your most recent trip.

Of course, all measures of fuel efficiency completely ignore inefficiency resulting from redundant routes, using a car when it's unnecessary, not taking public transport, not carpooling, etc.

Posted by Mike Hall | December 23, 2007 3:36 PM

> If you care about efficiency, the logical
> units are something like the reciprocal:
> either volume / distance, or, even better,
> volume / time. I

The problem is the computer would have to know your destination(s) to do that, and once programmed you would have to go there. Garmin aircraft GPS units can do this so I am sure their high-end auto GPSs could too but they probably don't activate the feature because only the most fanatic driver would ever program and reprogram the destination information.

Cranky

Posted by Cranky Observer | December 23, 2007 3:53 PM

It's worse than you say, Matt -- the person with the 100MPG car might drive more miles than the person with the 50MPG (or 18MPG) car, because it's cheaper to do so.

Posted by JSE | December 23, 2007 3:57 PM

Of course, the genius of pricing carbon through a tax or through auctioned emissions permits is, once again, that is spares people the burden of trying to do all the math in our heads and just lets price signals automatically find the most economical way of reaching the targets.

How do price signals work in absence of some sort of explicit or implicit calculation? A lot of classical economics relies on the assumption that everyone does have a calculator running, for example that a change from a 35% marginal rate to a 33% marginal rate will have real world impacts on investment decisions. This just shoves the burden of doing the math from the conscious to the unconscious level and in my view is just another piece of economic voodoo. Markets are not magic and they certainly are not invisible magic working through "pricing signals" working "automatically" to find "the most economical way". Very little of our current market is automatic, just about everything is being shaped by conscious activity, which is why you can be a Marketing Major without having much math ability at all, still less an invisible magical calculator.

Milton Friedman would be proud. Me, not so much.

Posted by Bruce Webb | December 23, 2007 4:06 PM

How very kafaesque...

...citing me for "total complete b.s." and then, as evidence, PROVING my point, which was this:

It is not difficult difficult to dramatically improve fuel efficiency - cars companies have been making great strides in making engines more efficient. But instead of applying those gains to better mileage, car companies applied them to make bigger cars with better acceleration.

kafka comes along and tells me this is total complete b.s. Why? Because the big 3 are making fuel efficient cars for Europe - which sounds surprisingly like exactly what I was saying - that they *can* if they choose, make much more fuel efficient cars very easily.

Posted by McKingford | December 23, 2007 4:30 PM

This is an interesting example of a certain kind of misleading statistic. Matt concludes that "upgrading to 100 mpg can't save that much gas" in spite of the fact that the 100-mpg car is using 683 gallons (about $2100 at today's prices) less than the 18-mpg car.

The proper place for this kind of statistic is an eighth-grade math class (or maybe fifth-grade what with kids being so much smarter these days), but in any real discussion about CAFE or energy policy it's just a foul ball.

Probably the average American gets about 22 mpg today. Changing our import restrictions would make about 50 cars worldwide available in our market that get over 50 mpg. So a real question might be, if we dropped our trade barriers and offered government financing to people who wanted to buy a new car, how much could we reduce our carbon footprint? (Not to mention the stimulus to the economy of having money to spend on something other than gas.)

I'm guessing that, using off-the-shelf technology, we could reduce our energy use a good deal- if we wanted to. YMMV.

Posted by serial catowner | December 23, 2007 4:48 PM

First, re: immigration, if your metric is "minimizing gross domestic consumption", yeah, it makes a difference, but from a "stop global warming" standpoint - I mean, seriously, you go down to Mexico, it's not like they're all riding burros or something, they drive cars and pickups too.

I wouldn't want to see SUVs banned, just taxed or otherwise made expensive enough that people who don't need them buy something else instead (hey, remember station wagons?) SUVs weren't invented in the 1990s, it's just that that's when they began to become so popular. But before that they were seen for what they should be seen as: a specialty vehicle, useful for forest rangers, lifeguards, geologists, etc.

You do realize that the fall of the station wagon and the rise of the SUV was a direct consequence of CAFE standards, right? When standards were first introduced in the 1970s and ramped up into the mid-1980s, it wasn't really possible to build a station wagon that both met standards and had adequate performance and then sell it at a viable price point. Thus the rise of the minivan, which fell under the more lenient "light truck" standard, originally intended for vehicles aimed at rough rural and semi-commercial users. And they were kinda fuel efficient! But the performance was mostly just sufficient, and no one, no men especially, really wanted to think of themselves as a minivan driver, so as fuel shock memories faded, we got bigger engines and restyled bodies and off-road marketing, thus the SUV.

Posted by Senescent | December 23, 2007 5:24 PM

MPG is the equivalent of low-fat food. If you are eating low-fat potato chips all day, you still ain't healthy. If you drive your 40 mpg Prius all over town looking for organic filters for your coffee machine and put 30,000 miles on it in the processs you're a far bigger problem than your grandmother who drives 1,500 miles a year in her '73 Dodge Dart.

Gas taxes shouldn't be flat, they should progressive based on consumption of gasoline.

Posted by joejoejoe | December 23, 2007 6:42 PM

I didn't note specifically that the "light truck" loophole needs to be scrapped, but incorporating them into CAFE would be one way to limit their production. I'm actually more in favor of a carbon tax or cap-and-trade system that would put all vehicles on some sort of continuum rather than arbitrarily drawing a line to say which vehicles do or don't count. Or, for that matter, which ones can and can't be made. I have no problem with SUVs being available for those who need them, but they should be priced in accordance with their externalities.

Posted by Adam Villani | December 23, 2007 7:47 PM

I don't understand Matt's point:

"In policy terms, meanwhile, the upshot is simply that it makes more sense to focus on raising the efficiency of the least-efficient vehicles than creating new super-cars."

But we are talking about mandating new cars. If we mandate new cars that average 50mpg we are going to save more gas than if we mandate new cars that average 18 mpg.

Posted by cw | December 23, 2007 7:52 PM

> It is not difficult difficult to dramatically
> improve fuel efficiency - cars companies have been
> making great strides in making engines more
> efficient. But instead of applying those gains to
> better mileage, car companies applied them to make
> bigger cars with better acceleration.
>
> kafka comes along and tells me this is total
> complete b.s.

Peter Egan, who writes for Road & Track and is extremely knowledgeable about the auto industry and auto/engine design, had a column a few months ago discussing exactly this point. Of cars in production under the same name for the last 30 years most have incredibly higher horsepower and the same or somewhat lower mpg than the historical average. The Corvette has reduced weight, improved performance, and tripled fuel economy.

So yeah, your point is right. You can see the same thing from emissions testing results of a 2004 car vs. a 1984: the new car produces about 1/20 the pollutants of the older one, which is another indicator of increased efficiency. Heck, you can tell just by looking at the horsepower: a lot of SUVs have increased their hp by 100 (and even 200) since the _2000_ model year!

Cranky

Posted by Cranky Observer | December 23, 2007 8:21 PM

"kafka comes along and tells me this is total complete b.s. Why? Because the big 3 are making fuel efficient cars for Europe - which sounds surprisingly like exactly what I was saying - that they *can* if they choose, make much more fuel efficient cars very easily."

Manufactures don't choose which cars to make, consumers choose which cars to buy. Confronted with low (high) gas prices U.S (European) consumers buy large (small) cars with low (high) MPG.

Is there something difficult about this?

Posted by kafka | December 23, 2007 8:22 PM

> Manufactures don't choose which
> cars to make, consumers choose
> which cars to buy.

Hmmmm - that means the US automaking corporations are wasting $5-$10 billion/year on unnecessary marketing campaigns. I assume you have already filed a shareholder lawsuit to recover these monies?

Cranky

Posted by Cranky Observer | December 23, 2007 8:24 PM

It's worse than you say, Matt -- the person with the 100MPG car might drive more miles than the person with the 50MPG (or 18MPG) car, because it's cheaper to do so.

Speaking of fallacies...

I see this come up every time we talk about increasing fuel efficiency standards, and it pretends that the only factor influencing fuel consumption in vehicles is cost.

Suppose you take a 15 MPG car and make new models 45 MPG. A person buying a new model is a "threat" to drive three times as much...but that's presuming that they now all of a sudden have three times as much time to drive and the requisite interest to do so.

The person in the 15 MPG car spends, let's say, 2.5 hours a week driving to and from work and another incidental 3 hours a week driving for shopping/travel/etc. That same person, in order to use the same amount of gas in the new model, would have to spend 16.5 hours a week - more than a full waking day - driving. It would be a major lifestyle change that there simply isn't the economic impetus to affect.

The argument also comes to the tendency to take longer trips - but those longer trips still require time and interest. Better gas mileage won't give people more vacation time or put a shorter distance between point A and point B.

Posted by Starscream | December 23, 2007 9:06 PM

OK so it's not relevant to the policy, but I for one was very interested in the 15-to-18 being a greater change than the 50-to-100.

If I'm doing the numbers right, changing from 15 MPG to 18 MPG is also a greater change in total gas usage than going from a 100 MPG car to riding a bicycle (or driving a magical car that uses no gas at all). Interesting stuff.

What it means for policy is, for Christ's sake let's get those low-mileage cars off the freaking roads. We'd do way better w/r/t total gas usage if we had a national car mix of 100% Camrys than a fleet of 50% Explorers and 50% Priuses. 100% Camrys might even beat 50% Explorers and 50% bicycles.

Posted by Kent | December 23, 2007 9:21 PM

"Hmmmm - that means the US automaking corporations are wasting $5-$10 billion/year on unnecessary marketing campaigns. I assume you have already filed a shareholder lawsuit to recover these monies?"

I don't own stock in the big 3 but you should if you think they can dictate choices. After all, it goes without saying these marketing campaigns prevent people like you from going down to your local Honda, Toyota,(whatever) dealer and buying a car TODAY that meets or exceeds the CAFE requirements that Pelosi & Co. have proposed for (get this) the year 2020.

Oh wait...

Posted by kafka | December 23, 2007 10:20 PM

"You save more fuel switching from a 15 to 18 mpg car than switching from a 50 to 100 mpg car."

Is this a problem he thinks many people actually have?

Posted by Jinchi | December 23, 2007 10:20 PM

In policy terms, meanwhile, the upshot is simply that it makes more sense to focus on raising the efficiency of the least-efficient vehicles than creating new super-cars.

What Eric and Andrew are forgetting is that the vehicles with the lowest MPG aren't SUVs, they're semi trucks hauling cargo across the country.

These vehicles don't get 15 miles per gallon, they get about 5. We can't simply outlaw those vehicles, they'll require better engines, better aerodynamics or gas-alternatives. Tanks, planes, ships and other military vehicles are also susceptible to our dependence on oil and gas. Invent the 100 MPG passenger car and you've probably solved a lot of the problems troubling those other vehicles.

Posted by Jinchi | December 23, 2007 10:39 PM

The problem is the computer would have to know your destination(s) to do that, and once programmed you would have to go there. Garmin aircraft GPS units can do this so I am sure their high-end auto GPSs could too but they probably don't activate the feature because only the most fanatic driver would ever program and reprogram the destination information.

Cranky

Not at all. The car already gives instantaneous miles per gallon. If you know miles per gallon, then

gallons per mile = 1 / miles per gallon.

gallons per hour = gallons per mile * miles per hour

It's one line of code. I think the Priuses with the fancy displays can do this, even.

Assuming the car can measure distance, time, and fuel used, it's easy to give a quick wrap-up of the statistics for each trip, from ignition to when you pull out the key.

Posted by Mike Hall | December 23, 2007 11:44 PM

So, to do things in different units, in gallons per mile the difference between 15 MPG and 18 MPG is

1/15 - 1/18 = .06667 - .05556 = .01111 gal/mile

whereas the difference between 50 MPG and 100 MPG is

1/50 - 1/100 = .02 - .01 = .01 gal /mile.

Posted by Mike Hall | December 23, 2007 11:53 PM

Look, what we have here is a very elementary observation about proportions. I already knew darling Andy was functionally innumerate. I guess it's mildly interesting that Sullivan's even more of a dunce than I had him down for, but not by much. This feels like one of those I gotta pimp the other Atlantic scribblers content-free posts.

Posted by sglover | December 24, 2007 1:19 AM

kafka,

It seems you are looking to pick a fight when there isn't much disagreement. My point, which you seem to accept, is that the car companies "can" - as in, "have the ability" - to pretty easily make vehicles much more fuel efficient than they currently are; as you point out, they are *already* making these cars, but happen to sell them in Europe.

You seem to believe that the automakers are prisoners of the US market, which demands bigger, faster, stronger - rather than high mileage cars.

I certainly acknowledge that it is easier to market high mileage cars where the price of gas is higher. But you have to wonder why the Big 3 fight so hard against higher CAFE standards. Part of their argument is premised in the notion that it is difficult to make their cars more fuel efficient - a point we both know to be untrue. Why are they doing this - and thus being misleading?

I think the clear answer is that it is more profitable to sell bigger and faster cars than reliable high mileage cars. So the "market", which you believe forces the Big 3 into building less efficient cars (which ignores, incidentally, the fact that the non-Big 3 have been making huge gains - to the point where the very survival of the Big 3 is now in question - have made their gains by selling better and more fuel efficient cars) is a product of their marketing.

Posted by McKingford | December 24, 2007 8:15 AM

I think Mike Hall is spot on with inverting the units. It's much more sensible to say that your car needs x gallons for 100 miles than y miles per gallon. A 20 mpg car needs 5 gallons per 100 miles, for example. 25mpg needs only 4. Puts the emphasis on the gallons, not on the miles.

Posted by bs23 | December 24, 2007 11:00 AM

Two easy fixes:

1. Post all mileage as gallons per 100 miles. Then, the fuel waste as you go down the chart becomes more obvious. 15 MPG = 6.7 gallons/100 miles. 18 MPG = 5.5 gallons/100 miles. 25 MPG = 4 gallons/100 miles. 30 MPG = 3.3 gallons/100 miles.

I think the average person doesn't necessarily understand that an increase from 15 to 18 MPG saves over a gallon of gas every 100 miles.

2. Regulate weight. All else being equal, the lower weight car uses less gas. You could tax weight directly or, as I've proposed, tax a combination of high-performance and weight. The latter is the better course because sometimes heavier cars (like carpool vans) are more efficient on a gallon per person per mile basis and other measures of efficiency. (A full 11-passenger van getting 11 MPG gets 121 person miles per gallon. That's just a little better than three people in a 40 MPG car.)

Posted by Sean | December 24, 2007 1:22 PM

Two easy fixes:

1. Post all mileage as gallons per 100 miles. Then, the fuel waste as you go down the chart becomes more obvious. 15 MPG = 6.7 gallons/100 miles. 18 MPG = 5.5 gallons/100 miles. 25 MPG = 4 gallons/100 miles. 30 MPG = 3.3 gallons/100 miles.

I think the average person doesn't necessarily understand that an increase from 15 to 18 MPG saves over a gallon of gas every 100 miles.

2. Regulate weight. All else being equal, the lower weight car uses less gas. You could tax weight directly or, as I've proposed, tax a combination of high-performance and weight. The latter is the better course because sometimes heavier cars (like carpool vans) are more efficient on a gallon per person per mile basis and other measures of efficiency. (A full 11-passenger van getting 11 MPG gets 121 person miles per gallon. That's just a little better than three people in a 40 MPG car.)

Posted by Sean | December 24, 2007 1:23 PM

Contrary to what Sean proposes, it is not necessary to tax a complex combination of automotive performance and weight in order to reduce gasoline consumption. If you want to reduce gasoline consumption, tax gasoline.
If some genius invents a heavy, high performance car that still gets great mileage why should it be taxed more than a lightweight, low performing car that gets lousy mileage?
There's no need to make the gas tax code as complicated as the income tax code. Just tax the thing you're trying to discourage (gas consumption), and leave the rest alone.

Posted by Don | December 24, 2007 5:43 PM