MPGe is a Lie

The Environmental Protection Agency's (EPA) Miles Per Gallon Gasoline Equivalent (MPGe) "measurement" is how electric and hybrid vehicles compare "fuel efficiency". The rest of the world measures things differently, but we'll talk about that more later.

What's strange, is that some really basic math suggests the EPA's MPGe numbers deceptively misrepresent an electric vehicle's true "fuel efficiency" by at least a factor of two.

Sounds like crazy talk to me...but let's just look at some numbers and see what happens.

Measuring The Numbers

I'm going to start by measuring "fuel efficiency" as how much CO2 is emitted per unit of work produced. Measuring efficiency by how much waste/exhaust is created, per unit of fuel consumed and work produced, is fair enough. Right?

After all, EU nations, China and Japan all use grams of CO2 per kilometer (or per mile) as their metrics, so why not here in the USA? Why this silly "MPGe" number?

Obviously, a nation of EVs running on 100% solar isn't ever going to be realistic, we need to work with the Real World's numbers, as they are today. Which means we're going to use math that's straight from a May 7, 2010 joint EPA and Department of Transportation (DOT) rulemaking:

  1. For every 1 kilowatt hour (kWh) of electric power generated for the US power grid, 6.89551 × 10-4 metric tons of CO2 are also generated. That's equal to 689.551 grams, or 1.52 pounds, of CO2.

  2. For every 1 gallon of gasoline consumed, 8,887 grams of CO2 are generated. That's roughly 19.6 pounds. The EIA seems to also agree.

The Vehicles

Today, we're going to pick on the Chevy Volt, and Chevy Cruze, with the Volt being only 700lbs heavier (batteries!), both built upon the same GM Delta II platform and with nearly matching physical dimensions and drag coefficients (Cruze=0.298cd, Volt=0.287cd), this should resemble a reasonably fair comparison.


  • Cruze: 28 City / 42 HWY
  • Volt: ...

Well, we already have a little issue. As GM only publishes a composite score of 37 gas / 93 "electric" for the Volt.

That's alright though, the EPA's composite score just assumes 55% city and 45% highway usage, and lumps the two measurements into a single 37 MPG number, if we break things back out, the Volt ends up with 35 City / 40 HWY. Considering that the Volt and Cruze are very similar vehicles (with exception to weight), this makes sense.

Now, about the Chevy Volt's claim to 93 "electric" miles per "gallon"...


The EPA says that MPGe is a measure of the average distance traveled per unit of energy consumed. For years, we've been trained to seek the highest possible MPG for our gas/diesel vehicles, so with MPGe, we'll of course want to seek the highest possible MPGe as possible.

Or do we?

According to the EPA's formula, 33.7 kWh of electricity should be equivalent to one gallon of gasoline.

So, let's look at that for a moment.

The EPA says that producing 1 kWh of electricity also results in the production of 689.551 grams of CO2.

33.7 kWh * 689.551 grams = 23,237.8687 grams of CO2

Let's review the numbers again:

  1. The EPA says that the "gasoline gallon equivalent" in electricity produces 23,237.8687 grams of CO2

  2. The EPA also says one gallon of gasoline consumed produces 8,887 grams of CO2.

So, which one is it? 23,237 is nowhere near close to 8,887, yet apparently the same amount of "input" is producing grossly disperse outputs. The laws of physics don't allow for this.

The Real MPGe

On one hand, the EPA says that a 1 MPG car will produce 8,887 grams of CO2. On the other hand, the EPA says that a 1 MPGe car will produce 23,237 grams of CO2

We can't have it both ways.

If we adjust our 1 MPGe numbers, so that the car produces only 8,887 grams of CO2, then we have to downrate the Chevy Volt's claim of 93 MPGe by a factor of 2.615. Which means that, running in electric-only mode, a Chevy Volt's true gasoline-equivalent milage is 35.56 MPGe.

Strangely-enough, the Volt's gas-only 37 MPG rating is almost a dead-on match for its true electric-only 35.56 MPGe rating. It also appears to remain a very close match to the Cruze's gas-only MPG ratings, which again, only makes sense.

So, other than for baseless marketing spin, why does the MPGe number exist anyway? To confuse you? To mislead you into thinking you're getting something for nothing? To make you feel better about shelling out $35,000 for a car you could have had at 2/3rd that price?

Even GM Agrees

Now, here's the really interesting part. Even GM and the EPA would appear to agree that the MPGe numbers are grossly misleading.

Here's how.

Read the fine print. In particular, you're looking for text just to the right of the 93 MPGe listing. The text says "...36 kWh per 100 miles..."

36 kWh * 689.551 grams = 24,823 grams of CO2 is produced per 100 miles of travel, on electric only. We've already discussed how this isn't realistic, so let's peel the smoke and mirrors back.

24,823 grams / 8,887 grams = 2.79

2.79 is the factor by which MPGe overstates true fuel economy. So, now we want true fuel economy.

100 miles / 2.79 = 35.84 MPG.

There you have it. Even GM agrees that the MPGe number is grossly misleading. But if you don't do the math yourself, you'll be tricked into thinking your car gets 93 MPGe, when it instead performs at nothing even remotely close to that.

Conclusion & Followups

The facts are abundant here: MPGe is a grossly overinflated number that wildly misrepresents electric vehicle energy efficiency. There's no such thing as a mainstream commercially produced 93 MPG, or 93 MPGe, full-sized least, not yet.

I propose we stop fooling ourselves; that we drop this silly MPGe nonsense, and instead promote energy efficiency numbers that focus on what our final goal is: reducing CO2 emissions from tailpipes and chimneys. We would do this by using an easy-to-understand, global CO2 emissions standard, measured simply in grams per kilometer or per mile.

For even more technical followup and details, I encourage you to read articles such as: