The agency thinks the new approach will result in more realistic ratings, especially for hybrids.

The Environmental Protection Agency is getting tired of discovering automakers overstating their vehicles' fuel economy figures, and it's apparently actually trying to do something about it. The government regulator has issued a new proposal that would force companies to conduct road tests to calculate their figures. The regulation isn't a guarantee yet, though, because it first has to go through a period of public comment.

According to The Wall Street Journal, the proposed rules come partially as a result of Ford, Hyundai and Kia re-stating the fuel economy results of select vehicles after finding their original ratings to be inaccurate. Under the regulation, automakers would be required to create their calculations for air resistance and rolling friction from real-world data gathered on a test track, rather than from computer models. Currently, EPA test procedures take place on dynamometers and involve professional drivers executing standardized tests designed to mimic typical city and highway drives.

The agency thinks that the new approach will result in more realistic ratings, especially for hybrids. "Some automakers already do this, but we are establishing a regulatory requirement for all automakers," said Chris Grundler, director of the EPA's Office of Transportation and Air Quality, to the WSJ.

Ford faced a very similar problem with bad data in June when it re-rated the fuel economy of several – predominately hybrid – models downward. The company was using a metric called Total Road Load Horsepower that was a measure of resistance on the dynamometer during testing. However, for these vehicles, the TRLH value proved inaccurate and threw off the mpg numbers.

The EPA wasn't happy about the change. "This issue highlights the need for continued strong oversight of the fuel economy labeling program," said a statement from Grundler after Ford's announcement. "Consumers need to trust that fuel economy window stickers are giving consumers reliable and fair estimates of real world fuel economy."

There may also be good news for the fuel economy ratings of diesel vehicles in the future. According to the WSJ, Grundler is also tasking engineers to find out why the EPA's tests tend to rate diesels lower than their real-world results.

Do you think real-world fuel economy road tests are a good idea, or will they just lead to more variation? Vote in our poll below, then have your say in Comments.



I'm reporting this comment as:

Reported comments and users are reviewed by Autoblog staff 24 hours a day, seven days a week to determine whether they violate Community Guideline. Accounts are penalized for Community Guidelines violations and serious or repeated violations can lead to account termination.


    • 1 Second Ago
  • 63 Comments
      EbolaVirus
      • 5 Months Ago
      EPA testing is EXCELLENT for comparing vehicle A vs vehicle B. It is not >>great<< at estimating real world fuel economy, and is heavily technology dependent.
      Jim
      • 5 Months Ago
      Build a national test track that all manufacturers do their final mpg testing on. It doesn't have to be government owned, it could be owned and run by an independent private third party. Make the city and highway testing protocol universal across all manufacturers. This protocol would identify minimum speeds, distances, etc. so that every car essentially takes the same test. Sure, real world results would vary from these results, but at least all models would start from the same baseline.
        Greg
        • 5 Months Ago
        @Jim
        Jim, it's already a standard protocol. And Consumer Reports already does exactly what you suggest. The problem is that automakers aren't following the protocol, and when discovered, have to change their ratings. This is about enforcement, not reinventing the wheel.
      spaze47
      • 5 Months Ago
      this should have been done years ago instead of a climate controlled room.
      johnnythemoney
      • 5 Months Ago
      I'd say yes but there is no way real world testing is going to be equal for all cars and tests, the same way sometimes you need 10 minutes to get to work, sometimes it's 15. If some mean average testing procedure could be produced (like mileage calculated out of 10 tests in 10 different days but with a given range of meteorological conditions), it could work, but that's no easy or cheap task.
      superlightv12
      • 5 Months Ago
      I've found that the EPA estimates are pretty accurate for the cars I have owned. Gas mileage is going to vary considerably based on how you drive. Maybe they should give an estimated gas mileage range, like 22 - 26 mpg in the city.
        Joeviocoe
        • 5 Months Ago
        @superlightv12
        Do you own a car that has a claimed 40+ MPG highway? The problem is that automakers have learned to optimize the cars for the lenient highway test. People drive faster on the highway than the test.
        Spec
        • 5 Months Ago
        @superlightv12
        The City number is OK, the highway number is a bit of a joke because has an average speed of less than 50 mph . . . substantially slower than people drive on highways today.
          montoym
          • 5 Months Ago
          @Spec
          So, when you drive on the highway, you get into your car and are immediately transported onto the roadway at 65+mph then? There's no driving to the highway, accelerating, merging, passing, slowing down, nothing? Just all straight 65+mph driving huh? If you took a look at your average speed of your daily commute, I bet you'd be quite surprised at how low that figure actually is, even if much of your time is spent at highway speeds. Don't be confused by the average speed they use, it's simply that, an average.
      BryNum
      • 5 Months Ago
      Finally the government has some sense! It wasn't Obama's doing for sure!
      Ashton
      • 5 Months Ago
      I will not say YES to that questionnaire, because I don't want even more of my tax dollars stolen from me, and future generations to pay for an even bigger/always expanding government.
        Spec
        • 5 Months Ago
        @Ashton
        Yep . . . that way it goes . . . a small adjustment to MPG ratings and the next thing you know you are living in a North Korean authoritarian dystopia.
      nick.hunkar
      • 5 Months Ago
      The european testing method gives much more accurate data than our method. While testing on the road can give some real world numbers there are too many variables for a good comparison.
        b.rn
        • 5 Months Ago
        @nick.hunkar
        "The european testing method gives much more accurate data than our method. " Any basis for that claim?
        Greg
        • 5 Months Ago
        @nick.hunkar
        Absolute load of crap. The European tests permit far more cheating than the EPA's, and their numbers are routinely 20%-30% higher than real life.
      Jarda
      • 5 Months Ago
      Go away EPA, no one needs you.
      jphyundai
      • 5 Months Ago
      EPA mpg testing is a waste of taxpayer money. Just delete the requirement altogether.
      ed_rc
      • 5 Months Ago
      All the people that replied yes have no idea of the variability in "real World" testing. Add to that individuals variation and you can't compare one vehicle to another.
        jtav2002
        • 5 Months Ago
        @ed_rc
        And how is that any different than current testing. There are variables here too. Same with performance testing.
          b.rn
          • 5 Months Ago
          @jtav2002
          There are variables in any test. "real world" testing has substantially more variables.
      Naturenut99
      • 5 Months Ago
      There is nothing wrong with the current testing model. Other than more retesting by the EPA to make sure the company didn't cheat. I have zero problems getting the ePA rated mpg from the past couple cars. People should understand that hard acceleration and doing 80 on the highway you are not going to get the rated mpg.
    • Load More Comments