Ford Begins Production of V-10 Hydrogen-Fueled Engines
Delhi Nixes Two-Stroke Three-Wheelers; Conversion to Four-Stroke Mandatory

EPA: US MY2006 Light Duty Vehicle Fuel Economy Same as 2005

LDV fuel economy 1975-2006. Click to enlarge.

Model year 2006 vehicles offer an average estimated fuel economy rating of 21.0 mpg. This average is the same as last year and in the middle of the 20.6 to 21.4 mpg range that has occurred for the past fifteen years, and five percent below the 1987 to 1988 peak of 22.1 mpg, according to the EPA’s just-released report Light-Duty Automotive Technology and Fuel Economy Trends: 1975 Through 2006.

MY2006 light-duty vehicles are estimated, on average, to be the heaviest, fastest and most powerful vehicles than in any year since EPA began compiling such data.

Since 1975, the fuel economy of the combined car and light truck fleet has moved through four distinct phases:

  1. A rapid increase from 1975 to the mid-1980s;
  2. A slow increase extending into the late 1980s;
  3. A decline from the peak in the late 1980s until the mid-1990s; and
  4. A period of relatively constant overall fleet fuel economy.

After two decades of constant growth, light truck market share has been relatively stable for five years, and represents about 50% of the total market. Growth in the light truck market was primarily due to the increase in the market share of SUVs, which increased their share by more than a factor of ten, from less than two percent of the overall new light-duty vehicle market in 1975 to more than 25% of vehicles built each year since 2002.

The increased market share of light trucks, which in recent years have averaged more than 6 mpg less than cars, accounted for much of the decline in fuel economy of the overall new light-duty vehicle fleet from the peak that occurred in 1987-88.

Weight versus acceleration. Click to enlarge.

Improved engine, transmission, and powertrain technologies continue to penetrate the new light-duty vehicle fleet.

The trend has clearly been to apply these innovative technologies to accommodate increases in average new vehicle weight, power, and performance while maintaining a relatively constant level of fuel economy.

At the peak of industry-wide fuel economy in 1987, some automaker groups had average fuel economies 6 to 8 mpg higher than others.

Average fuel economy in 1975, 1987 (the peak for fuel economy) and 2006 for the top 8 automakers in the US. Click to enlarge.

In MY 2006, the gap has decreased to a maximum difference between automakers of an estimated 5 mpg, with a typical difference between higher and lower fuel economy marketing groups being 3 to 4 mpg.

For MY2006, Honda, Toyota, Hyundai, and Volkswagen all have estimated fuel economies of 23.5 to 24.2 mpg, while General Motors, Nissan, Ford, and DaimlerChrysler all have estimated fuel economies of 19.1 to 20.5 mpg. Each of these groups has lower average fuel economy today than in 1987.

The fuel economy values in the report are based on “real world” estimates provided by the Federal government to consumers and are about 15% lower than the values used by manufacturers and the Department of Transportation (DOT) for compliance with the Corporate Average Fuel Economy (CAFE) program.

Because it has been more than two decades since the current procedures for determining real world fuel economy estimates were established and because both vehicle technology and vehicle driving patterns have changed, EPA has proposed changes to the methodology for calculating real world fuel economy estimates and expects to finalize a new methodology by the end of 2006.




An average of averages?

When determining the fuel economy, is this simply (A + B + C)/3 for a manufacturer with three models, or is it (A*numA+B*numB+C*numC)/(numA+numB+numC), or something else? How are these averages calculated?


It is so called “fleet average” which you described in second formula.

fyi CO2

"EPA has proposed changes to the methodology for calculating real world fuel economy estimates and expects to finalize a new methodology by the end of 2006."

After 20+ years, we might really have to take the bags off our heads?


Of course for the most part MY2006 vehicles were sold in 2005 and early 2006 when the gas prices slacked off a bit and everyone was grabbing employee pricing and 0% interest financing on SUVs from domestic manufacturers.

Nissan's VQ35DE sure has brought down their fuel economy shoving a 2.5L engine in the little Sentra. Once upon a time the Sentra was almost close to the Corolla and Civic as far as gas mileage is concerned.


Notice VW's fuel economy is almost the same now as it was in 1987...also notice that they have the fewest number of SUV/Truck choices for sale (Toureg and that Eurovan).


It is so called “fleet average” which you described in second formula.

But this too is an average of averages, and therefore not an accurate representation. I shall use an example to demonstrate:

Vehicle A: 10 mpg, sell 1 unit
Vehicle B: 20 mpg, sell 4 units
Vehicle C: 50 mpg, sell 10 units

So, using the fleet average you're looking at:
(10*1 + 20*4 + 50*10)/15 = 39.3 mpg.

But, that's an average of averages. Let's say each driver goes 10,000 miles per year. We can calculate how many gallons will be needed, divide, and then know the true fleet mpg.

Vehicle A: 1000 gallons * 1 vehicle == 1000 gallons.
Vehicle B: 500 gallons * 4 vehicles == 2000 gallons.
Vehicle C: 200 gallons * 10 vehicles == 2000 gallons.

Total gallons: 1000 + 2000 + 2000 == 5000 gallons.
Total miles: 10,000 * 15 = 150,000.
Total miles per gallon: 150,000/5000 = 30 mpg.

So, why do we get two different numbers? Because we're averaging averages in the first example, something that is sure to create problems because you're adding numerators and adding denominators, without finding like terms. The second example in this post is the correct measure of the fleet mpg, assuming that all vehicles are driven the same number of miles.

So, I ask again: what formula is used?


Quote (from "All average fuel economy values werecalculated using harmonic, rather than arithmetic, averaging."

There is more explicit information if you follow the link cited at the bottom of the article.

Rafael Seidl

Carmakers have been, to some extent, simply responding to legislative demands for higher safety and customer demand for more space and comfort. There's well over 100lbs of acoustic damping material in a modern car, for instance (though some of it serves multiple purposes).

Until 2003, fuel economy was largely a non-issue for US consumers. Prices were low because US fuel taxes are low. For the same reason, the increase in oil prices since then has led to a much higher increase in prices at the pump, in relative terms, than in Europe. It is this rate of increase, at least as much as level per se, that tends to send people into a tizzy.

If oil prices stay high and/or the US were to raise fuel taxes, the present consumer demand trend toward smaller, more fuel-efficient vehicles will persist. Any carmaker unwilling to see the writing on the wall will go bust.

Changing the EPA guidelines on sticker MPG is welcome in that in will reinforce this trend, regardless of what happens to CAFE and the gas guzzler tax.

The litmus test will come if and when oil prices come down again, due to expanded production in e.g. Canada plus political stabilization in the Middle East. Granted, lower oil prices are not in the offing right now because global demand is strong, but past experience suggests that supply will catch up again eventually. The oil business has gone through many booms ans busts.

You might not see $1 gasoline ever again, but $2 is conceivable if GHG emissions continue to be blithely ignored in Washington, DC. If and when that happens, will US consumers continue to eschew SUVs? Human nature suggests that sadly, this will not be the case unless fuel prices stay high (and go even higher), regardless of what happens to the price of crude oil. Breaking an addiction means not going back to the drug pusher.



Past experience may be inoperative in this case if you believe people like Simmons and most of the posters over at The Oil Drum.

In any event, we shouldn't take the chance that they are wrong. As someone I read said recently, we use a very high discount rate for the future, too high for our own good.

Broken record here, but both you and I agree that we need to implement taxes, phased in if necessary. The basic difference is that I think they need to be phased in at a much higher rate and need to go to at least $7.00 per gallon.

Of course, this will never happen, especially in an election year. We like to believe in magic, unlimited ethanol and the free lunch we associate with hydrogen. If things get really tough, we can always take over another country --- say Iran. After all, Iraq has turned out so well and the oil from that country is paying for the war. Also, the predicted decreases in oil prices because of that conflict are coming along swimmingly. Doncha know.

Oh, I forgot. Israel may do the Iranian job for us. Let's hope they are willing to share the oil.

Harvey D.

t: Many of your assumptions may be coming soon. We have not yet realized that energy conservation and cleaner, higher efficiency vehicle will become a neccessity. We still believe that alternative fuels will be plentiful and cheap and that our VUS and 4 x 4 dinosaurs will be bigger, faster and have 500+ HP for centuries to come.

My pick-up is bigger and faster than yours mentality/brain washing is still very present. A $7/gal to $10/gal at the pump would help to de-programme the majority. Politicians, for well known reasons, will never do it. The only alternative is to drive crude oil price (3X or more) from $78/barrel to $234+/barrel. A three fold increase has been seen before and will happen again.

Oil producing/exporting countries like, Iran, Venuezela, Mexico, Saudi Arabia, Irak, Canada, Kuwait, UAE, Algeria etc will reap it in, if they are smart enough to collect and not let the huge profits-revenues in the hands of oil industries.

Many countries now realize that rare non-renewables like natural gas and crude oil should benefit (mainly) the country and not so much the foreign or local oil industries.

It will be interesting to see who will end up with ownership of the future alternative fuel industries. To ensure long lasting revenues, crude oil producing countries could re-invest part of the huge oil profits into US and European alternative fuels and energies.


Stormy: The intention is to reduce fuel consumption (Gallons per mile), so a harmonic average is used.

Average fleet fuel economy = Sum (Model sales)/Sum(model sales/model fuel economy)

So in your example, it would be 30 MPG

Leszek Pawlowicz

Changes in the EPA methodology don't mean changes in the computation of CAFE. When the EPA modified the sticker mileage numbers in the 1980s to make them match real-world numbers better (dropping the city by 8% and the highway by 22%), they were only allowed to do so on the condition that the new numbers not be used in the computation of the CAFE. It's likely to be the same with any new EPA methodology used to calculate more realistic real-world mileages. Only Congress and the Executive Branch have the power to modify the CAFE standards, and they don't seem particularly inclined to.


Hopefully, with surging oil/gas prices the lower EPA numbers on new car stickers will make more consumers ask for a slightly more efficient vehicle if given two nearly identical choices.

tom deplume

The CAFE formula could be even more misleading because of the assumption that each class of vehicle put on equal mile per year. Most commercial vehicles have been large SUVs and pickups which put on 2 or 3 times as many miles per year than most small cars. Such effects could double true fleet average but not be reflected in CAFE figures.

Sid Hoffman

Good point Tom. For that matter, the specialty makes like Ferrari and Lamborghini are extra unfairly punished via CAFE even though owners typically put only about a tenth as many miles on a high dollar sports car as they put on a family sedan. But yeah, point being that in a perfect world, you'd need to make it mandatory to report mileage of every vehicle in annual vehicle inspections, and then report that mileage back to the EPA to calculate your CAFE score.


Those three graphs are shameful...whose worse the mfgs, consumers or our nannies?


As i know, niche car brands like Lamborghini and Ferrari are exempted from CAFE standards.


fred wrote: "whose worse the mfgs, consumers or our nannies?"

If the manufacturers cared about anything beyond the current quarter's numbers, or if the consumers as a whole cared about anyone but themselves (and were decently informed regarding the consequences of their decisions), then we wouldn't need "nannies", would we? But they don't, and they don't and aren't. It's a curious situation that the nannies are bought and paid for by the "children" (the mfgs) that they are supposed to be minding. If the "nannies" worked for the consumers, we wouldn't have the "light" truck mess that we have today.

The comments to this entry are closed.