Fuel economy follies: Cheatin' or mistaken?

Automotive journalists have been hearing a consistent message from Korean automakers Hyundai and Kia: their attributes top the competition. Over the last few years, they have claimed superiority for most of their products in nearly every measure that matters, from power and torque to interior and cargo capacity to, most importantly, fuel economy.

Hyundai logoWhen Hyundai introduced its new compact Elantra a couple years ago, execs poked fun at rivals for using asterisks to designate that their 40-mpg compacts – vehicles like the Chevrolet Cruze ECO, Ford Focus SFE, Dodge Dart Aero and Honda Civic HF – were special high-efficiency models. The Elantra and friends, they boasted, needed no such disclaimer because every model was EPA rated at 40 mpg highway. Yet they also claimed better power, torque and performance.

As a recovering engineer, I have wondered how that could be possible. How can nearly every Korean car and crossover boast better power, torque and fuel efficiency than every other in its segment? Could Korean engineers be that much smarter than everyone else? They certainly are smart, and hard working. But so are Japanese, Germans and Americans.

When I've questioned such "better at everything" claims at Hyundai and Kia new-product introductions, I've received little response beyond smiles and smugly shrugged shoulders. But US auto engineers – who routinely test, analyze and benchmark competitive products – have told me (off the record) that, in their own testing of Hyundai and Kia vehicles, they've been unable to achieve the Koreans' advertised numbers.

Wouldn't it be embarrassing and image damaging for them to get caught cheating?

Could it be, I wondered, that Hyundai and Kia have been fudging their fuel-economy numbers, especially those very important 40-mpg EPA highway claims? Doesn't our all-powerful EPA audit and check automakers' claims to keep them honest? Wouldn't it be embarrassing and image damaging for them to get caught cheating? Yes, sort of and yes.

But how many Americans recall that these same Korean companies were caught a decade ago inflating their power and performance claims? Very few apparently remember, or care, since both have enjoyed record-setting US sales since then. But cheating on EPA fuel economy ratings? How could they do that, and how have they gotten away with it?

But now, as most readers know, Hyundai and Kia (which is owned by Hyundai and shares its platforms and powertrains) got caught doing exactly that by the EPA, which started auditing their fuel-economy claims following consumer (and competitor) complaints. And these audits have shown that they have "overstated" (by one, two or more miles per gallon) the fuel-economy ratings of nearly a million 2011-2013 vehicles sold through October, 2012.



Red-faced Hyundai/Kia executives blamed "procedural errors" in testing in Korea. They apologized, killed their "40-mpg" ads and began putting corrected window stickers on unsold vehicles. They are also compensating (via debit cards) owners of affected vehicles for the differences between claimed and corrected gas mileage. And, of course, at least one class-action lawsuit has been filed against them by legal vultures looking to cash in.

My theory is that the easiest way for an automaker to inflate mpg ratings would be to deflate the "road load" power curves.

My theory has been that the easiest, and most difficult to catch, way for an automaker to inflate its fuel-economy ratings would be by deflating the "road load" power curves it uses to calibrate dynamometers (stationary vehicle test-lab treadmills) for EPA highway emissions/fuel economy tests. Because dynos operate indoors, these road-load numbers are needed to simulate aerodynamic drag and other factors that affect a vehicle's real-world energy usage.

A vehicle's road-load power requirements – the energy it needs to roll, throttle off, at a constant speed – is determined by very precise coast-down testing. According to the SAE procedure, these tests are run with the vehicle coasting in neutral on a long (typically two-mile) stretch of dry, clean, straight, perfectly-level road, at air temperatures between 41 and 95 degrees F, with little wind and no precipitation. A minimum of 10 runs (five each direction) is required, each beginning at a minimum of 125 kilometers per hour (77.7 miles per hour) and ending after the vehicle has coasted down to 15 kph (9.3 mph). Data is continuously recorded between 115 and 15 kph.

With no suitable facility to run their own coast-down tests, the EPA has had to accept automakers' road-load power numbers.

Through complex calculations, this data then determines a vehicle's road-load energy usage as a function of speed, which becomes a major mathematical factor in the EPA's highway test calculation. So if its reported road-load curve is inaccurately low, a car's EPA highway rating (and also, as a result, its city/highway "combined" rating) will be unrealistically high. And while the EPA audits automakers' test procedures and runs emissions/fuel-economy tests (on a small percentage of vehicles) in its own Ann Arbor, MI dynamometer lab to validate their results, with no suitable facility to run their own coast-down tests, they have had to accept automakers' road-load power numbers as submitted.

And, sure enough, Hyundai/Kia says the "procedural errors" that led to its bogus fuel-economy claims involved coast-down testing. Is it possible that their engineers misinterpreted the very specific SAE procedural instructions and/or miscalculated the road-load numbers derived from their coast-down data? Possible, yes. But not if they're the world's smartest.

2012 Hyundai Elantra window sticker

Now that the Koreans have been caught with their hands in the fuel-economy cookie jar, the EPA, the media and class-action parasites are also going after Ford's impressively high (and impossible to achieve in the real-world) 47 mpg city/47 highway/47 combined ratings for its Fusion and C-Max hybrids. Ford doesn't help its credibility by routinely stating its vehicle's highway ratings as if they were real-world – but so far is standing by its hybrid mpg numbers.

Kia logo To see which current vehicles come closest to achieving their (usually realistic) EPA combined ratings in real-world driving, I sorted my recent test-vehicle data in order of the mathematical differences between them. Most vehicles from most makers have come pretty close. But topping the list of overachievers was the 2012 Mercedes-Benz E350 Bluetec hybrid I drove back in May, 2012 at 33.2 mpg observed vs. its 25-mpg rating. Four other 2012 German cars (a VW Toureg, a Mercedes S550, an Audi TT RS and an Audi Q7) were next in line at four-to-seven mpg over their EPA combined numbers.

Notable at the bottom of my chart were a 2013 Ford C-Max Energi plug-in hybrid at 32.4 mpg, a whopping 14.6 below its 47-mpg combined rating, and a regular C-Max at 34.2 mpg, although (to be fair) they were driven in November/December cold weather. A 2013 Ford Fusion Hybrid (October) also under-performed at 35.6 mpg (11.4 mpg below), while a Toyota Prius C (April) came in at -10.2 mpg, a Kia Soul (March) at -8.5, a Toyota Camry Hybrid (September) at -5.3, a Hyundai Elantra at -4.3 and a Hyundai Genesis Coupe at -3.9 mpg.

Because I drive each car (as an owner would) in different weather and road conditions, some mostly on freeways, others more on urban and suburban roads, depending on my schedule, these are interesting but unscientific comparisons. And I should add that any damage done to Hyundai's and Kia's brand images by this recent inflated-fuel-economy-claims embarrassment seems fleeting so far, since both have continued their hot sales unabated.

Share This Photo X