Bringing Reality to Gas Mileage Figures

By Cindy Skrzycki
Tuesday, May 24, 2005

It's another kind of sticker shock.

What new car owner hasn't found, after some real-world driving and a fill-up at the pump, that the sticker in the window promising decent gas mileage was considerably inflated?

Now, for the first time in 30 years, the Environmental Protection Agency may have to get in gear to revise the test it has been using on new vehicles to determine a more realistic estimate of expected mileage for city and highway driving.

That's because, with gas prices at high levels, consumer and environmental groups backed a provision offered by Sen. Maria Cantwell (D-Wash.) in the just-passed Senate highway bill that would require the EPA to issue a new rule to bring the test results closer to the real world.

The EPA first started testing vehicles in 1974 to gauge their expected miles-per-gallon performance. In 1985, the agency "adjusted" the numbers downward, but hasn't addressed further how changes in technology, models and driving habits affect gas mileage. The agency says it has been analyzing the gap for the past couple of years and was already planning to address it.

But Cantwell said her amendment was necessary to make sure the agency takes action. Changing the test may result in estimates that are 10 percent to 30 percent lower than consumers are used to seeing on a new-car window label, according to supporters of the legislation and the Energy Department , which tracks annual energy use nationwide.

"We want consumers to believe in this number and the information," Cantwell said in an interview. "EPA needs to do a new test. It's hard to be accurate if you don't base the test on the driving behavior and conditions we have today."

The provision calls for the agency to take into account speed limits, acceleration rates, braking, weather variations, vehicle load, use of air conditioning and driving patterns when it comes up with a fuel economy number. If the measure is accepted by a House-Senate conference committee, the EPA will have until the end of the year to issue a proposal. The final rule deadline is 18 months later.

Currently, car manufacturers test prototypes and then the EPA checks their accuracy at its lab in Ann Arbor, Mich. The vehicles are tested by placing their wheels on a dynamometer, a kind of indoor treadmill that simulates driving. A professional driver takes the vehicle though driving routines. The amount of carbon in the exhaust is measured to determine the amount of fuel used.

The problem is the conditions under which the test is done: The speed is an average of 48 miles per hour, with a top speed of 60. There is no air conditioner creating a gas mileage "penalty." There is no road congestion or accelerating to high speed.

"If you drove that way, you would get great gas mileage," said Chris Plaushin , national manager of regulatory affairs for AAA . "But there would be an angry line of people behind you."

To give its members a more realistic idea of what to expect, AAA does its annual mileage tests on as many as 200 different vehicles. It puts kids in the back, groceries in the trunk, and drives in both city and highway conditions, rush hour, stop-and-go, with the AC running and the windows down.

CONTINUED     1        >

© 2005 The Washington Post Company