Drivers fooled by 'MPG illusion'

[nivi]

http://www.guardian.co.uk/environment/2008/jun/20/travelandtransport.carbonemissions

Car manufacturers should change the way they advertise the fuel efficiency of their vehicles, to make the environmental impact of buying a new car easier to understand, researchers said yesterday.

Turning a car's fuel efficiency on its head by stating it in "gallons per mile", instead of "miles per gallon" helps potential customers choose the greenest option when they upgrade, said the team at Duke University in North Carolina.

In The MPG Illusion, published in the US journal Science yesterday, Richard Larrick and Jack Soll at the university's Fuqua School of Business describe how the use of miles per gallon is misleading and causes people to grossly misjudge the environmental impact of upgrading to a new car.

The two management professors stumbled across the problem while working out the true fuel efficiency of different cars in a car-sharing scheme. They found people often believed - mistakenly - that a 10mpg improvement in fuel efficiency always corresponded to the same fuel saving.

"The reality that few people appreciate is that improving fuel efficiency from 10 to 20mpg is actually a more significant saving than improving from 25mpg to 50mpg for the same distance of driving," said Larrick.

Likewise, replacing a car that does 10mpg with one that appears only slightly more efficient at 11mpg saves as much fuel as upgrading from a 33mpg car to a 50mpg car.

To work out new fuel efficiencies, Larrick calculated how many gallons of petrol a car would burn over 10,000 miles. A 10mpg car would burn 1,000 gallons, while a 20mpg car would burn only 500 over the distance. In comparison, a 25mpg car would use up 400 gallons, while a 50mpg car would use 200. Despite most people's first impressions, the fuel saving is 2.5 times greater in the first instance.

The finding suggests that people who use two different cars equally will always save more on fuel by improving the least efficient car.

In a series of experiments, Larrick and Soll asked volunteers to study a series of cars whose fuel efficiencies were given in miles per gallon. When they were asked which upgrade would save the most fuel, they invariably made the wrong decision.

In one test, most people believed that upgrading a car from 34mpg to 50mpg would save more petrol than replacing an 18mpg car with a 28mpg vehicle. In fact, the latter saves twice as much fuel.

"Miles per gallon is misleading and can play tricks on our intuitions," said Soll.

"It made us realise that low-efficiency cars really use a tonne of gas and drive overall consumption and that's why even small changes in low mpg cars makes a big difference," said Larrick.

"Changing the way we express efficiency would help the car companies make clear to buyers where there are gains to be made."

When the tests were repeated using gallons per 100 miles, the volunteers correctly picked the greenest option from those available.

"There are significant savings to be had by improving efficiency by even 2 or 3mpg on inefficient cars, but because we communicate in miles per gallon, that saving is not immediately evident to consumers," said Soll.

The authors call on car manufacturers and consumer publications to list efficiencies in terms of gallons per 10,000 miles driven, making it easier to see how much petrol a car might use in a typical year of driving, and how much fuel and money could be saved by opting for a more efficient car.

According to the US department of energy, a Prius can cover 48 city miles per gallon, while a Mercedes Benz G55 SUV, will cover only 11 miles on the same amount.

David Bonilla, an energy expert at Oxford University's transport studies unit, said the move was along the lines of other countries, which already state fuel efficiencies in litres per kilometre, and EU plans to describe vehicle efficiencies in terms of the amount of carbon dioxide released per kilometre.

"Fuel efficiency isn't a top priority for most people when they are buying a new car, but that's beginning to change because of high oil prices and environmental awareness," he said.

"Sales of cars with low fuel efficiencies have already dropped in the US and there's evidence over the past year that people are shifting to more fuel efficient vehicles. We can expect that to happen here too."

Way to go guys, I admit, I was fooled too.

 

[ArneHD]

Ok, can someone please explain this too me? I don't get it.

Edit: He isn't talking relative change, is he? In relative terms both are the same.

 

[Huayna Capac357]

@ ArneHD:

Let's say 5 different cars drive 1,000 miles. One car gets 10 mpg, one gets 20, one gets 30, one gets 40, and one gets 50. Thus, the first car would require 100 gallons of gas, the next one 50, the next one 33, the next one 25, and the next one 20. Most people would think that going from a 30 mpg car to a 50 mpg car would save mre fuel than going from a 10 mpg to a 20 mpg. The difference in gallons for the former example is only 13 gallons (33-20= 13), while the latter example saves 50 gallons (100-50=50). Using a more fuel effecient car is still greener, but when you change from a your car (x) to a new one 10 mpg more, it will be less significant if x is a higher number.

 

[bhsup]

What kind of BS is this?

"Likewise, replacing a car that does 10mpg with one that appears only slightly more efficient at 11mpg saves as much fuel as upgrading from a 33mpg car to a 50mpg car." ??????

If I have to travel 500 miles:
10MPG = 50 gallons
11MPG = 45.45 gallons

33MPG = 15.15 gallons
50MPG = 10.0 gallons

So 10 to 11 saves me 4.55 gallons and 33 to 50 saves me 5.15 gallons. I call the claim rubbish. Oh, and not to mention 33MPG saves 30 freaking gallons over the lower numbers to begin with.

 

[Echse]

I think that [unit of volume] per 100 [unit of length] is surely easier to understand especially for the people not able to do simple math.
And to be honest I never would have thought that anyone does this vice versa.

Are all cars in the US sold with miles per gallon?
But I guess this shouldn´t matter because if you grown up with a system like that you should got used to it when your old enough to buy a car.

The much bigger problem at least in Europe is that a car's fuel efficiency is measured in a lab under perfect circumstances which never can be found in real life.

 

[bhsup]

The US Government has new standards for 2008 that are supposed to be more realistic. Great site.

http://www.fueleconomy.gov/

 

[IglooDame]

So they think I'm doing a greater good by driving my ~24mpg Mustang instead of my ~19mpg Escape, moreso than the fellow who drives his Prius instead of his Honda Accord.

 

[skadistic]

We need a smiley of a head hanging down and shaking its back and forth in disappointment.

There should be a one semester mandatory class in highschool on automotive learning. Just to teach the simple things like how to change your oil and figure out what is the most fuel efficient car.

 

[LightFang]

Wow, I was fooled. That was very interesting.

But now that I think about it, gallons per mile would make far more sense...and would be easier in terms of math, methinks.

 

[emzie]

The litre / 100 KM thing is done here, but I still see most people use MPG.

As to fuel efficiency, while the study is true, it misses the mark that the 50 mpg car is going to be cheaper to run than the 11 mpg SUV. Always. The amount saved by switching from a 10mpg SUV to an 11 mpg SUV might be more, but you still pay more over all than the guy in the Prius. Which is what matters to most people at the end of the day.

 

[Ofuh]

Wait, what's the point of this article?
I hope I'm missing something if all I got from it was that it is basically saying if you add 3 to 10 it is a smaller increase than if you add 1 to 2, which is pretty common sense...

 

[Brighteye]

It's because mpg is an inverted number. But it seems like a bit of a niche area. The values work when demonstrating what is best: more miles to the gallon. Choosing which car of two to upgrade seems like a secondary consideration, after buying efficient cars in the first place.

 

[bhsup]

...
As to fuel efficiency, while the study is true...

But that's the problem, Augurey, it isn't true.

What kind of BS is this?

"Likewise, replacing a car that does 10mpg with one that appears only slightly more efficient at 11mpg saves as much fuel as upgrading from a 33mpg car to a 50mpg car." ??????

If I have to travel 500 miles:
10MPG = 50 gallons
11MPG = 45.45 gallons

33MPG = 15.15 gallons
50MPG = 10.0 gallons

So 10 to 11 saves me 4.55 gallons and 33 to 50 saves me 5.15 gallons. I call the claim rubbish. Oh, and not to mention 33MPG saves 30 freaking gallons over the lower numbers to begin with.

Also, honestly, what is the difference between 30MPG and 3.3 Gallons/100 Miles? They are the same thing, and 30MPG says it quicker.

 

[uppi]

Wait, what's the point of this article?
I hope I'm missing something if all I got from it was that it is basically saying if you add 3 to 10 it is a smaller increase than if you add 1 to 2, which is pretty common sense...

The point of the article is, that the reciprocal nature of mpg is counter-intuitive and might lead to wrong decisions.

Let's say a family has two cars: one small with 30 mpg, and a big one with 10 mpg. They want to save fuel costs and are planning to buy a new car. However they can only afford one new car, so they have to choose which one to upgrade. They are considering two replacements: A big one at 20 mpg and a small one at 50 mpg. They might think, that the 20 mpg savings of the small car are better, so they might go buy that one. However, the 20 mpg car will result in better savings (assuming the usage of both cars is about the same)

Now if the efficiency was given in liters per 100km (or gallons per 100 miles if you want to continue using outdated units...) it would be instantly clear, that going from 20l /100km to 10 l/ 100 km is better than going from 8l /100km to 5l/100 km.

 

[Masquerouge]

I'm really confused now :crazy:

 

[skadistic]

A car that gets 33mpg is better then one that gets 30. A car that gets 42mpg is better then one that gets 40.

 

[Mirc]

And I thought it was some illusion about the .MPG (.MPEG) file format.

 

[Masquerouge]

A car that gets 33mpg is better then one that gets 30. A car that gets 42mpg is better then one that gets 40.

But if I understood the article right, it's better to go from 10 to 20MPG than go from 25 to 50...

To work out new fuel efficiencies, Larrick calculated how many gallons of petrol a car would burn over 10,000 miles. A 10mpg car would burn 1,000 gallons, while a 20mpg car would burn only 500 over the distance. In comparison, a 25mpg car would use up 400 gallons, while a 50mpg car would use 200. Despite most people's first impressions, the fuel saving is 2.5 times greater in the first instance.

VRWC, what do you make of those numbers? They seem legit to me.

 

[bhsup]

Yeah, those are legitimate for that instance. I was calling foul on the ridiculous 10 to 11 vs 33 to 50 claim.

 

[Aegis]

I actually heard this on the radio a few weeks back by an announcer trying to get people to feel better about driving gas-guzzling vehicles:

The announcer said, "Studies have proven that under the right conditions, a large SUV may actually have better gas consumption than a smaller compact car! Studies show that SUV's driven for 6,000 miles a year use the same as, or less, gasoline than compact cars driven 14,000 miles a year."



Yeah, and if you drove that compact car only 6,000 miles, you'd be saving more gas and money.