In this era of pricey gasoline, fuel-efficient cars are getting attention. A subcompact Chevy Sonic that gets 40 miles per gallon? Intriguing! But many drivers have found that a car’s advertised mileage is often quite different from how it actually performs on the road. So how big a problem is this?

Better double-check that sticker. (Honda/WIECK)

Part of the problem, the study found, was that the European system for testing cars — known as the NEDC — doesn’t do a very good job of simulating real-world conditions. The E.U.’s regultaors test cars at low accelerations and at constant speeds, which isn’t how real-life people drive vehicles. What’s more, cars are tested without the air conditioning running — and A/C is a giant fuel-guzzler. For all these reasons, a German car’s official fuel-economy rating (measured by how much carbon-dioxide it releases) has become increasingly detached from reality.

The authors of the ICCT paper, Peter Mock, John German, Anup Bandivadekar and Iddo Riemersma, warn that European automakers are at risk of losing credibility with consumers — and could even end up hurting the market for new, fuel-efficient cars. (Right now, the EU is trying to reduce greenhouse gas emissions from the transport sector 70 percent below 2008 levels by 2050.)

Okay, but that’s Europe. How big a problem is misleading mileage ratings here in the United States? I asked Jim Kliesch, a vehicle expert at the Union of Concerned Scientists, to walk me through this.

The first thing to note is that different U.S. agencies use different tests to determine the mileage of the car. There’s the testing procedure that the Department of Transportation uses to figure out a car’s mileage under the Corporate Average Fuel Economy rules. This test was first set up in the 1970s, and it’s awfully archaic. The cars are basically run on giant treadmills, and the test was designed for less-powerful cars of a bygone era that lacked air conditioning. Basically, a modern car’s CAFE rating is going to be much, much higher than the fuel-economy that the car actually gets in the real world.

But there’s also the mileage sticker on a new vehicle. This sticker is crafted by the Environmental Protection Agency, which uses much more modern and accurate testing to rate a vehicle. The EPA takes into account city driving and highway driving, it looks at how cars operate in cold conditions and with air conditioning. The agency recently revised its tests in 2008 after responding to criticisms of its methods.

And the EPA’s numbers are far different. To give a sense, the Honda Civic gets 40 miles per gallon according to government CAFE ratings. But in the showroom, the official sticker only says it gets 29 miles per gallon — this is the rating that accords to the EPA’s more accurate test. (Kliesch has written up a policy brief that helps people translate CAFE standards into on-the-road mileage.)

But how accurate are the EPA’s mileage stickers? It depends. “Today’s modern vehicles are very sensitive to the way we drive them and the specific features they contain,” says Kliesch. “So driving aggressively can you give a much lower fuel economy than the window sticker indicates. Conversely, if you drive gently and don’t drive at excessive speeds, you can actually do a lot better [than the sticker mileage]. A lot of it is in the hands of the driver.”

By and large, experts argue that the EPA does a pretty decent job with its tests: Popular Mechanics recently tested two cars rated 40 miles per gallon, the Hyundai Elantra and Ford Focus, and found that the cars actually performed slightly better in the real world than the EPA’s ratings suggested. Still, a few disgruntled American car owners have occasionally sued automakers like Honda and Hyundai for allegedly misrepresenting the performance of their vehicles.

So the United States doesn’t seem to have Europe’s problems with misleading mileage stickers. Still, as gas keeps getting pricier and every gallon counts, this is likely to be a contentious issue anyway.