« Gasoline Price Gouging? | Main | Girth vs. Growth »

September 08, 2005

For Fuel Economy, the Numbers DO Lie

With gas prices soaring, some people may trade in their gas-guzzlers for more fuel efficient vehicles. But don't trust the EPA ratings. A recent analysis by Consumer Reports shows that 90% of vehicles get worse gas mileage than advertised -- in some cases more than 50% worse for city driving. And nationally, Corporate Average Fuel Economy (CAFE) may be overstated by a whopping 30 percent.

How does this happen? Manufacturers inflate fuel efficiency in several ways. First, they don't test cars the way people actually drive. Vehicles are tested in a laboratory, not on actual roads. And while the EPA assumes that 55% of driving is done in city traffic, which uses more fuel than highway driving, many cars actually spend 62% of time there, according to Consumer Reports. Second, the car they test is not the car you buy. Manufacturers are allowed to use prototypes built especially for the fuel economy test, so they often modify them (within limits) to get the best rating possible.

In 1984 the EPA responded to an outcry by consumers who were angry that they could not get the fuel efficiency advertised. But rather than change the way it tests cars, the EPA just adjusted the test results it reports: 10% lower mpg for city driving, 22% lower for highway. And, as Consumer Reports shows, even these adjusted numbers still aren't accurate, especially in city driving.

And worse, CAFE standards, already low and full of loopholes, are also affected. Automakers successfully lobbied so that only the unadjusted mpg ratings are used when enforcing CAFE standards. While the government estimates that the fleet of 2003-model-year passenger cars that Consumer Reports tested averaged 29.7 mpg, Consumer Reports only got 22.7, well below the current standard of 27.5 mpg. For light truck the difference was 21.4 mpg versus 16, with a standard of 20.7 mpg.

When buying a new car, follow Consumer Reports's advice:

The EPA sticker can help you evaluate relative gas mileage among vehicles, but not absolute mpg.. .. [D]iscount the EPA sticker numbers for city travel as follows: conventional cars and trucks, 30 percent; larger hybrids, 35 percent; diesels, 36 percent; smaller hybrids, 42 percent.

For more information, go to the sourceConsumer Reports has the complete article online plus its rating for 303 vehicles (use links in their left column), available until November 2, 2005.

Posted by Jessica Branom-Zwick | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference For Fuel Economy, the Numbers DO Lie:


Consumer Reports has had screwed-up methodology in its testing of hybrid cars, but won't admit it. They continue to claim that hybrids get only about half of the rated mileage, when the vast majority of hybrid drivers report no such problems.

I drive a hybrid, primarily in the city, and don't pay particular attention to the various mileage-maximizing tricks some drivers use (e.g., daily tire inflation check/top-up, driving below the speed limit, etc.). My mileage is lower than EPA estimate, but by 15%, not 42%. That is, on city driving, I get about 42mpg rather than 48mpg.

Based on the comments on the multiple hybrid mailing lists and websites I read, my experience is typical.

I'm sure CR is right that the real-world use doesn't match the EPA tests, but their numbers regarding hybrids are simply way, way off.

Posted by: Jamais Cascio | Sep 8, 2005 5:01:41 PM

ONE VEHICLE stands out among all others in consistently representing its true fuel efficiency: the bicycle! At zero gallons of gasoline per mile, no matter where you ride it, you can trust your legs--and your wallet--to tell you how truly efficient a bicyle is.

Posted by: Valerie Hahn | Sep 21, 2005 9:48:49 AM