By on September 4, 2011

There was troubling news at the end of last week, as Automotive News [sub]‘s Rick Kranz reported that an unnamed automaker was quietly accusing another unnamed automaker of tweaking its EPA fuel economy tests, arguing

There’s a noticeable difference between the mpg number posted on some cars’ window sticker and an analysis of the data submitted by automakers to the EPA.

We’ve tried to get several automakers to comment on the accusation, but nobody wants to touch it. But, as we’ve looked into the issue, a few more details have surfaced that seem worth sharing. Hit the jump for the latest…

With such a serious accusation floating around, it was inevitable that AN [sub] would revisit the story (although technically the entire thing has been reported on AN’s blog, distinguishing it from a normally-reported news item). And sure enough, an update was posted yesterday by David Guilford, who writes

Over lunch recently with a former product executive for a major automaker, I asked if he thought anyone was fudging their numbers.

He said that he doubted that a company would out-and-out cheat. But then he smiled and said that, well, there are ways to game the system.

For instance, he said, his former employer used to test 50 vehicles of a given model, knowing that a normal bell-shape distribution of results would probably produce one outlier with higher mpg. Results from that car would be reported.

But — here’s the kicker — the automaker decided to stop playing games. Not because other automakers complained. Not because the feds were applying pressure.

No, the automaker discovered that it was angering its customers, who complained that their mpg didn’t match the sticker.

As I commented when we first reported the story, there seems to be something of a gentlemen’s agreement not to report suspected “fudging” of reported fuel economy numbers, as a number of models have a reputation for failing to achieve their claimed numbers (Chevy’s Equinox and Hyundai’s Elantra are the most common examples). And this attempt to close the barn door only reinforces that impression. By trotting out a convenient storyline in which market functions solve the problem of unrealistic EPA numbers (which has already stopped anyway), the industry seems to be closing ranks to keep out any nosy federal investigators. After all, if a thorough verification of EPA numbers were undertaken, who knows where the black eyes might end. The story also confirms that cheating is probably not egregious, although if there are systematic discrepancies between window sticker numbers and reality, the feds should still take measures to restore consumer trust in the EPA’s numbers regardless of how those discrepancies got there.

Meanwhile, the main mystery of the story, the question of who precisely is accusing who, remains unanswered. And given Guileford’s conciliatory walk-back post (not to mention the industry’s considerable incentive to let this episode blow over), we’ll likely never know what the real story here is. But there’s a pretty clear consensus on the prime suspects, and like any good whodunnit, the investigation begins with a look at motive. Ford and Hyundai have been very publicly feuding for the fuel economy leadership halo, making them something like the beneficiaries of a freshly-changed will in an Agatha Christie novel. In fact, a look back at AN [sub]‘s blog archives yields yet more evidence that could shed light on a number of aspects of this mystery.

AN [sub]‘s cub reporter assigned to Ford (and Porsche’s dream reviewer), Jamie LaRue set the stage for Kranz’s piece over a week ago, when she blogged:

Hyundai Motors Co. promotes that its redesigned Elantra earns 40 mpg on the highway. This must vex Ford Motor Co.

Ford’s redesigned Focus gets 38 mpg highway — unless a buyer opts for the SFE package (40 mpg).

Since the arrival of CEO Alan Mulally in 2006, Ford has set out to be the fuel economy leader in every segment. That’s why Hyundai’s claim must be especially irksome.

So by gosh, Ford is going to prove that it is the leader regardless of what window stickers or advertisements say.

Well, by gosh, how did they do it?

During a media drive event at Ford’s proving grounds in Romeo, Mich., this week, Ford engineers had each reporter drive the Elantra at 45 mph around a 2.5 mile course. The same reporter was then asked to drive the Focus (not the SFE model) at the same speed around the same course. All conditions with the cars were equal, except for the drivers. An engineer in the backseat of each car monitored the fuel economy each earned.

After dozens of reporters drove the vehicles, the data was calculated and the results were in. The average fuel economy earned by our group driving the Focus was 40.4 mpg versus the Elantra’s average of 37.8 mpg.

Read all about it in Motor Trend’s December issue (published in October). Wait, hang on… what was this supposed to prove again? “Ford Focus: Optimized For 45 MPH Cruising” is not a particularly snappy tagline. Help us out, Jamie!

So Ford proved its point to a group of automotive journalists.

What did it accomplish? Not sure.

But, hey, I’m writing about it.

Actually Jamie, you’re blogging about it. Because writing those facts as a news item would get you a good laughing-at from all but the worst editors. Luckily there’s the old blog, where meaningless propaganda from the company you’re covering needn’t be judged by the harsh standards of “professional journalism.”

Speaking of which, is there anyone still reading this who doesn’t believe Ford is accusing Hyundai of manipulating EPA numbers? Outside of the accusing company itself, only the folks AN [sub] knows for sure what’s going on here… and they’re already walking back the story, regardless of how painfully obvious it is. But hey, thus far the story has yet to make the sacred leap from mere “blog item” to hallowed “news item,” so it doesn’t actually reflect on anyone’s credibility. Nobody even needs to ask Ford why they aren’t willing to put on some man pants and make a real accusation if they want to make an accusation. The whole story can be safely ignored now, as the professionals who started it all are ready to get back to some “real journalism.”

Get the latest TTAC e-Newsletter!

34 Comments on “Fuel Economy Fudge-Gate: An Update...”


  • avatar
    kol

    According to Fuelly, right now the Honda Elantra and Ford Focus are basically tied for reported fuel economy. Although the 2011 Ford Focus slightly beats them both.

    Consumer Reports tests show the Focus as having higher actual highway fuel economy, but lower actual city fuel economy, though the difference is small.

  • avatar
    Robert Fahey

    Seems the current system is akin to letting parents submit their kids’ SAT tests — with random spot checks only. Thank God crash tests aren’t handled this way:

    http://www.fueleconomy.gov/feg/how_tested.shtml

  • avatar
    PintoFan

    Eh. The entire thing has too much “truthiness” about it at this point to really interest me. The only real evidence that we have that EPA numbers are being “fudged” is purely anecdotal, based on “my cousin’s boyfriend’s sisters Equinox only got 25 MPG” type stories. If somebody can prove to me that the sticker numbers really are being “gamed” by at least 3-4 mpg based on hard data, then we can talk. If you treat EPA sticker as anything more than ballpark you’re kidding yourself anyway, there are way too many real-world variables that can change it.

    Could it be Ford accusing Hyundai? It’s in the realm of possibility. But it’s equally likely that it could be Toyota or Honda accusing Hyundai, or GM accusing Ford, or any other number of permutations. Because the Japanese have fallen behind in the MPG race (at least on paper), this could be one of them trying to gain a competitive edge by undermining the credibility of the competition.

    • 0 avatar
      CJinSD

      Did you read the story and miss the part about the Ford media stunt that revealed Ford to the be the company trying to expose Hyundai for gaming the EPA tests?

      • 0 avatar
        PintoFan

        I did see that. However, it proves nothing, other than that Ford wanted reporters to see that the Focus gets better mileage (under controlled conditions) than the Elantra. At best, it’s circumstantial evidence, at worst it’s an innocuous comparo being read as part of a much wider conspiracy that doesn’t actually exist. Your contempt of Ford is pretty well known so it doesn’t surprise me that you would be pushing Ed’s hypothesis.

    • 0 avatar
      Thinkin...

      Here’s a pinch of data for you:

      http://www.thetruthaboutcars.com/2010/02/why-the-chevy-equinox-epa-mileage-numbers-dont-add-up/

      Here’s the gist: Basically GM gave the Equinox the same EPA fuel ecomony rating as the Chevy Malibu. The cars partially share a platform, and are equipped with identical entine/transmission combinations. The Equinox is 500# heavier, and considerably less aerodynamic than the Malibu. (The Equinox has both a larger frontal area, and a higher c/d.)

      Thus, the laws of physics – as much as marketers like to bend them – simply won’t allow an Equinox to get the same fuel economy as a similarly powered Malibu. However, that’s what the EPA numbers say. This technical problem has proven true in the real world as well, where the Malibu achieves near its EPA numbers, and the Equinox doesn’t even come close.

      How the EPA would let such numbers pass muster is anybody’s guess. (Mine is that the government just happened to own GM during this debacle.)

      • 0 avatar
        redav

        That seems sound to me.

        I do think that certain models are gamed if there’s a critical value to hit, like the 40 mpg in the compact segment. It’s a psychological threshold, and it’s no surprise that every brand hit it at the same time using various strategies.

        Also, brands will publish hp, torque, towing, etc, values based on what their competitors say. I remember one case where (IIRC) Nissan leaked values which Ford matched, but the leaked values were deliberately low, so at the end of the day, Nissan had the higher numbers.

        So, long story short, I expect in cases where there’s no benefit from fudging the numbers, car makers don’t do it. But if they HAVE to match a competitor, they will.

  • avatar
    obbop

    Gasoline additives of various types.

    Consider the difference between “winter and summer gas.”

    There are other additives, some required by various statutes, decrees and by private desire such as Techron in Chevron, et al.

    Corn likker’ is well-known.

    Would the lack thereof of certain required additives or addition of other additives even if not commonly or never used but not patently illegal be used if their presence upped MPG figures?

    Just curious.

    Curious Coot.

  • avatar
    gslippy

    “After all, if a thorough verification of EPA numbers were undertaken, who knows where the black eyes might end.”

    Sounds just like the issue of steroid use in Major League Baseball.

    I’ll agree with PintoFan that with all the variables at play, even Ford’s comparison test leaves much to be debated. And we don’t even know who’s pointing at whom; it could be anyone, but in the case of Ford v Hyundai, maybe Hyundai is accusing Ford of fudging to get closer to its own numbers.

    BTW, 45 mph seems like a very low speed at which to hit EPA highway numbers; I would have expected at least 55 mph.

    • 0 avatar
      charly

      is 45mph even legal speed in the USA or drive cars always slower or faster

      • 0 avatar
        gslippy

        45 mph is a very common speed limit in many suburban/business areas, but it’s often combined with lots of stopping.

        A true highway EPA mpg test should be at a steady 55 mph, in my opinion. 65, 70, and 75 mph limits are restricted to non-urban areas, and although large, they are not as common as 55 mph.

      • 0 avatar
        aristurtle

        55 is also too low. Actual average speed of traffic is generally between +5 and +10 of the speed limit on most American highways (when there isn’t a traffic jam, anyway). Here in MD the speed cameras don’t even bother to take a picture until you’re 12 MPH over.

  • avatar
    Educator(of teachers)Dan

    Tell ya what. Give a given car that somebody thinks is cheating to me to drive for a week, then give the car to my mother. Average the two. (I can tell you that her numbers will be higher than EPA and mine will be lower.) :P

    It is sort of a non-story, manufacturers are trying to meet a government standard and avoid penalties for having a CAFE that’s too low. Yes if you game the system too hard you are going to piss off customers. Unless there is an official investigation it will all be hearsay.

  • avatar
    Rican5.0

    So wait, reported economy numbers are based on a type of honor system? Wow, silly me here I thought there was actual testing by a government agency. Would explain how my Genesis 2.0T has never even come close to the factory claimed MPG numbers.

  • avatar
    L'avventura

    So why then doesn’t an entrepreneurial journalist actually test these sticker claims using actual EPA test methods? We’ve seen frequent anecdotal experiences about mileage from automotive journalists, but never has there been an attempt to independently verify any of these claims.

    • 0 avatar
      kol

      Most in-depth reviews do include information about the gas mileage observed during the period the car was driven. Consumer Reports provides this sort of information for all the cars it tests including observed numbers, EPA numbers, and MPG during a 150 mile trip.

      I suspect no individual journalist does it because A) it is a huge undertaking and B) whoever is the loser will never give that journalist a press vehicle again.

      • 0 avatar
        L'avventura

        Currently these ‘in-depth’ information by publications like CR on mileage are completely anecdotal and unreproducible. 150 mile trip, or even a long-term test vehicle does not provide any validity to EPA claims. Your 150 mile trip is different from mine, and there are no controls in the test methodologies.

        The subject at hand here is that there are VERY serious accusations being made that some in the industry are fabricating their EPA numbers to some degree, and the industry as a whole is colluding to hide this fact. If this is true, it goes beyond deceiving the consumer, it also means that environmental and industrial laws that revolve around EPA mileage standards are miscalculated. Future CAFE regulations are based on measured improvement from the current EPA mileages, there are many billions of dollars at stake, if there is deception in this metric than the regulatory implications are significant.

        And yes, it is a big undertaking, especially doing it right. But then again large publications tend to do fairly extravagant overseas supercar showdowns these days. And yes, the exposition of deception in their EPA ratings will most definitely not make automotive companies happy. This either Pulitzer material or complete career suicide.

      • 0 avatar
        redav

        Honestly, there isn’t a need to replicate the EPA test methodology. All that is necessary is to compare cars’ ranking per EPA and per your own reproduceable tests.

        For example, CR tests the Focus & Elantra according to their own controled, rigorous test. If the Elantra has higher EPA numbers but lower CR numbers, something’s fishy.

        That to me is the real benefit of the ratings–not absolute values to plan my finances around–but as a means to compare various vehicles. If a vehicle has a higher EPA mpg, I expect it to get more mpg. For an extreme example: A Focus has higher EPA mpg than an F-150. If I drive both like a granny, the Focus will get better mpg. If I drive both like sports cars, the Focus will still get better mpg. This validates ranking based on the EPA numbers.

    • 0 avatar
      Barry K. Nathan

      The closest thing I can think of is the Consumer Reports fuel economy testing, which seems to be an alternative test procedure that tries to compete against the EPA tests. By the way, this is how CR feels about the EPA fuel economy tests.

  • avatar
    APaGttH

    I still think this is smoke where there is no fire.

    There are just so many darn variables. The average auto journo knows about as much about cars and driving as they do about brain surgery (there aren’t a lot of Jack Baruth’s of the world).

    In cars designed to squeeze every last drop of MPG out of them, variables like 2 PSI of air pressure in the tire, oil level in the crank case, summer versus winter blend, E10 versus pure gas, shoot, even the weight of the driver or the amount of equipment added for pictures and testing can have an impact. Never mind individual driving styles, ambient temps, tarmac surface conditions, traffic, timing of traffic lights, accessories run during evaluation, the list goes on.

    There is so much that can impact real world driving and conditions. I’ve always figured on the 2008+ EPA standard that if I get within 10% of the MPG +/- then I’m at least close.

  • avatar
    amca

    I point the finger at Hyundai. Their cars seem just a tad too efficient for their specifications. Just a hunch I’ve had for some time.

    • 0 avatar
      tedward

      agreed. Add to that the fact that no one has been reproducing the EPA numbers in tests and it’s pretty damn obvious. It really is an awful practice, especially when you consider how conservative EPA numbers have traditionally been, and are, for other manufacturers. The question usually is, “how much better than that am I going to actually do?”

      Everyone jacks their numbers, and soon, if this isn’t nipped in the bud publicly.

  • avatar
    NormSV650

    I’d figure Honda insight would be on that list too.

    Funny yopu look through Fuelly and most of the mpg figures are low side. But combined city/highway really close.

  • avatar
    tomLU86

    I like to think Consumer Reports is a good source for real-world fuel economy.

    But it’s not as good as it was. In the 70s, I remember CR would post city, and highway figures at 40, 50, and 60mph. Then in the 80s, they went with one highway figure.

    The 80s also marked the transition from a true ‘consumer’ publication to one that catered to the whims of the upper-middle class. So, if middle America liked luxurious cars, CR would evaluate them–no more (correct) comments about the cost of repairs.

    So, for example, when an 86 VW Jetta got 21 mpg city and 40 mpg highway, is that comparable to a 2011 Jetta getting 18 mpg city and 34 mpg highway? Or are the test different? CR doesn’t say..we are supposed to beleive them…

    My interest in mpg borders on the obsessive…every fill-up is recorded. Over the years, I’ve made some interesting observations. Of course, my driving has differed, but in general, I found that

    my 86 VW GTI, 1.8 liter 102hp, manual 5-spd fuel-injected, got overall 28-29 mpg, and on long trips 32-35 mpg depending on speed (but those speeds where in the 60-68 mph range–late 80s, early 90s). In suburban driving, it got 25-28 mpg.

    My 2011 Malibu, 2.4 liter, 6-spd auto, power everything but seats, A/C, has average 26 mpgs. But, half of those miles are long trips, where I’ve gotten 31-35 mpg at speeds of 66-73 mph. Around town, I’ve gotten 21-24 mpg.

    The Malibu is much bigger and more comfortable on a trip, and faster AND thriftier. But the mass takes it’s toll in suburban driving.

    What’s fascinating is the impact of mass. If ALL the feds regulated was tailpipe emissions and we had a free market, cars would weigh less. They could get by with smaller engines. With todays engine electronics, it’s possible that car sized like 70s Ford Fairmont or 80s Accord or 90s Civic would get 30 mpg in suburban driving, and be reasonably zippy.

    We have gotten spoiled by our engineers, who have harnessed technology to enable the internal combustion engine to deliver levels of power and efficiency that were unimaginable. My former 2008 Chevy Silverado extended cab with 4wd got better mileage than my mother’s 75 Pontiac Ventura 260-V8…and it weighed 2000 lbs more and had much quicker acceleration.

    Mass is the enemy…if the govt was serious about saving fuel, they’d FIRST: RESCIND all auto regulations except tailpipe emissions–on the principle that we all have to breathe the air. Then they’d raise gasoline taxes. Then we’d use less fuel, import less, and have cleaner air, AND a better standard of living. Lower-middle class working Americans who must commute far (because they can’t afford housing close to cities) are deprived of the choice new fuel-efficient cars, because they are too expensive–in part because they weigh to much and have too much stuff people don’t really need. Is that fair? How can liberals abide the 2nd class treatment of workers? How can conservative abide the lack of free choice?

    If you want crashworthiness or air bags or whatever, fine, the market will provide it. But if you don’t, the market can do that too..and you can save A LOT of fuel.

    I write too much, don’t I?

    • 0 avatar
      jeffzekas

      I totally agree with tomLU86: focus on emissions and efficiency, and, with the exception of seatbelts (they have saved LOTS of lives) stop stressing on every detail of automobile safety… AND put a $2 a gallon tax on gas (to be used for high speed Japanese-style rail systems)… then let the market do the rest.

    • 0 avatar
      geozinger

      I would largely agree with most of what you posted. Having been in a couple of accidents years ago, I’ve sustained a couple of injuries to my lower back, knees and neck.

      I too, drive a larger car now than I did back then, but I am grateful for the advances in safety gear.

      I believe if I’m involved in an accident again, my chances for injury would probably a lot less. I’d rather take the penalty of burning more fuel than take the penalty of another injury to my body.

    • 0 avatar
      aristurtle

      If you want crashworthiness or air bags or whatever, fine, the market will provide it.

      Except it actually won’t, because human beings don’t work that way. Look up “prospect theory”. In general people deliberately take more risks when it comes to avoiding a loss (not hurting themselves and others in a collision, for example) than they do when it comes to acquiring/maintaining a gain (e.g. saving money on a vehicle’s purchase price and fuel costs). This is why the last time you were in a car accident the other guy didn’t have insurance. The government is serious about saving fuel but it is also serious about trying to reduce the number of pointless auto accident deaths. Granted, there is generally too much envelope-pushing going on (I’ll accept airbags, but is a weight sensor in the passenger seat really necessary?).

      Besides, if you want reduced safety, low weight and fuel savings, buy a motorcycle. Few safety regulations, cheap, low weight, fuel economy ranges from “good” to “excellent”.

  • avatar
    Volt 230

    When was the last time someone drove between cities or town at 45 mph?. Even on secondary roads, the normal speed is 55 to 60.

    • 0 avatar
      mazder3

      Try New Hampshire some time. Even when the speed limit is posted at 50-55 most schmucks go 40-45. Makes me yearn for car-to-car missiles….

      • 0 avatar
        Volt 230

        Not here in Fla. I guess because it’s full or tourists and the roads are nice and open plus distances are pretty long.

      • 0 avatar
        mcs

        Try New Hampshire some time.
        Be sure to wait for foliage season for the maximum effect. That’s when it gets really bad.

      • 0 avatar
        mazder3

        Or try going from the Maine coast to the Monadnock Region on Labor Day. Gotta love Routes 4, 9 and 202. The main east-west corridor in NH is a two-laner with few opportunities to pass due to the massive amount of traffic. Still funny that the slowest bits (35mph in a 55) were on I-93 ’cause there was a Statie in the median and on I-95 over the Piscataqua River. F’n lookie-loos.

  • avatar
    philadlj

    This is NOT meant as a sarcastic stab at Chevy or GM, but I feel like it’s a pertinent question: has anyone out there been able to reach 32 mpg on the highway in an Equinox, or driven 600 miles on a single tank, as its initial marketing insisted?

    Here’s Ed’s Feb. ’10 article about it: http://www.thetruthaboutcars.com/2010/02/why-the-chevy-equinox-epa-mileage-numbers-dont-add-up/


Back to TopLeave a Reply

You must be logged in to post a comment.

Subscribe without commenting

Recent Comments

New Car Research

Get a Free Dealer Quote

Staff

  • Authors

  • Brendan McAleer, Canada
  • Marcelo De Vasconcellos, Brazil
  • Matthias Gasnier, Australia
  • W. Christian 'Mental' Ward, Abu Dhabi
  • Mark Stevenson, Canada
  • Faisal Ali Khan, India