Varun Sivaram

Energy, Security, and Climate

CFR experts examine the science and foreign policy surrounding climate change, energy, and nuclear security.

Print Print Cite Cite
Style: MLA APA Chicago Close


Do Gasoline Based Cars Really Use More Electricity than Electric Vehicles Do?

by Michael Levi
October 26, 2011

Business Insider published an interview today with Tesla founder Elon Musk in which Musk makes a striking claim: “You have enough electricity to power all the cars in the country if you stop refining gasoline,” he asserts. “You take an average of 5 kilowatt hours to refine [one gallon of] gasoline, something like the [Tesla] Model S can go 20 miles on 5 kilowatt hours.”

It’s a claim that I’ve seen pop up frequently in recent weeks, often framed as an assertion that electric cars use less electricity than normal ones. There’s only one problem: It’s not true.

The math behind the claim is simple. Refinery efficiency is about 90 percent and the energy content of a gallon of gasoline is about 132,000 Btu. Put that together and you have about 13,000 Btu of energy cost per gallon of gasoline produced, which is equivalent in energy terms to 4 kilowatt hours. (The 6 kilowatt hour claim is based on outdated efficiency figures.)

But this is flawed in two big ways. The first is that most of the energy used by refineries doesn’t come from electricity; only about 15 percent of it does. That cuts the electricity-used figure down to about 0.6 kilowatt hours. The second is that conversion of fuel to electricity is pretty inefficient. A process loss of energy equivalent to 0.6 kilowatt hours translates to an actual electricity loss of around 0.2 kilowatt hours.

This is consistent with refinery data. The Department of Energy estimates that refiners used 47 TWh of electricity in 2001 to produce refined products from 5.3 billion barrels of oil. Assuming  that you get 42 gallons of refined products from each barrel of oil, this works out to about 0.2 kilowatt hours of electricity used for each gallon of gasoline produced.

Post a Comment 12 Comments

  • Posted by J-M

    Ah yes, but have you considered the energy required to operate each and every gasoline/diesel pump in every gas station across the country? Plus the amount of energy to transport the fuel to stations? Taking into consideration the oil sands in Alberta, Canada (currently the single largest oil import of oil and gas to the US) energy required to extract then seperate the bitumen from the sands before sending it to get refined (often across the border in the US and even more so if Keystone XL gets the go-ahead. Plug those numbers through and Elon Musk may have been more accurate than you suggest.

    [ML: Nope. Refining dominates well-to-tanks energy use. Even if you doubled the energy use he’d still be way off.]

  • Posted by ANDY APPAN


  • Posted by Steve Jones

    There are a lot of numbers in this post, I think the “6 kilowatthour…” in the third paragraph should read “5 kilowatt hour…”.Shouldn’t it?

    [ML: Lots of different numbers out there. 6 kWh was the original; Musk rounded down.]

  • Posted by J-M

    So how much energy is required for every transaction at the pump and the transportation from well to tank? Daily, monthly, yearly averages. Has anyone ever calculated this?

  • Posted by Hans Nicolaisen

    Above Michael writes, “…the energy content of a gallon of gasoline is about 132,000 Btu.”

    This is the first time I’ve ever seen a figure that high for gasoline, and have always used 125,000 btu/gallon in my own calculations, since that seemed to be the general consensus. (The EIA figure for 2009 gasoline is 124,238.)

  • Posted by perk

    On the electric side — coal is producing 45% of the electricity for the car. Rail transportation costs of the coal are about 6% of net generation, and the average efficiency of about 33%, then you have line losses to house, recharger losses (15%?) battery losses, and electric motor and drive chain losses. Internal combustion results in losses along the chain that are in different places, but end up very ( + or – 10%) similar for similar size cars. So does the amount of fossil fuels used for both options.

    Really only a full calculation of the energy use beginning with raw material prospecting to measuring work done (kwh or jules) but the final cars, will tell you for sure.

    Today, electric cars mostly move the pollution from the cities to the power plants. When a substantial portion of our electricity is from sources other than coal or gas — things will be a bit different.

    [ML: All true. I was talking about something else.]

  • Posted by JP White

    The author of this articles claims the following.

    But this is flawed in two big ways. The first is that most of the energy used by refineries doesn’t come from electricity; only about 15 percent of it does.

    This raises the question, So the other 85% of energy used to refine Oil to gasoline. What form does that take? The author doesn’t say.

  • Posted by Rob

    JP White, the vast majority of the remainder comes from natural gas. Natural gas that could be producing electricity directly for charging EV’s, and skip the inefficient step of using it first to refine gasoline.
    Many studies show we could charge between 60-80% of the US vehicle fleet with surplus overnight generating capacity without building any new generating capacity. Then consider that around 40% of EV buyers thus far have also installed solar panels on their roof to offset their vehicles power use.

    So now we see that we can already charge most of the vehicles overnight, the vehicles plugged in during the day in homes and at workplaces can charge with additional solar panel output, 10-15% can be offset by electricity use previously used in refineries and if there’s any remaining increase in demand after all those offsets then some of the natural gas that provides 85% of the energy input used to refine gasoline can be diverted to power generation.

    Combine all these factors and all the fear mongering about how we are going to charge EV’s start to seem unfounded.

  • Posted by Eric Cha

    Hi Michael,

    Interesting article. While I can see where the accuracy of Mr. Musk’s claim is disputable, I can’t help but come away from your article feeling like it is somewhat of a red herring. As JP White points out in a previous comment, you neglect to say where the other 85% of the energy required to refine crude oil into gas comes from. The answer (according to the data source you link) is (mostly): natural gas, petroleum coke, and still gas (along with some other energy sources that are small in comparison). The fact of the matter is, the whole POINT of Mr. Musk’s statement was that a LOT of ENERGY is consumed just REFINING oil into gas. Perhaps that energy is not taken in the form of directly supplied electricity, but that does not change the fact that it is ENERGY, which could be used for other purposes, or as Mr. Musk points out, could be used to produce electricity which could then be used in EV’s which are far more efficient. The overall point that EV’s are far more efficient with energy usage ESPECIALLY when considering efficiency of production.

  • Posted by Jeff Chan

    “But this is flawed in two big ways. The first is that most of the energy used by refineries doesn’t come from electricity” is the most flawed thing about this piece. The energy must come from somewhere, and it must be counted as part of the energy cost of refining. Also if your numbers say 4 kwh to refine one gallon of gasoline and Elon’s say 5 kwh, that’s close enough in the big picture to say that the same energy could be used to propel an EV nearly as far as the refined gasoline, without actually using the gasoline. In other words gasoline is very inefficient especially if the processing energy is included.

  • Posted by Thomas

    You say:

    “The second is that conversion of fuel to electricity is pretty inefficient. A process loss of energy equivalent to 0.6 kilowatt hours translates to an actual electricity loss of around 0.2 kilowatt hours.”

    But i can’t follow your logic. Who’s process does this process loss relate to? I can’t find any way of reasoning that makes your application of this factor make sense.

    If the refinery uses 0.6 kWh of electricity, then that’s what it used. Full stop. No justification for applying a further conversion factor.

    Further, if the other 85% of the 4.5kWh came from using up fuel energy during the refining process, we should call that lost potential electricity — apply the 1/3 conversion factor to that and add it to the refinery’s effective electricity cost.

    In total, 4.5 x 0.15 + 4.5 x 0.85 x 1/3 = 0.675 + 1.275 = 1.95 kWh

    So even accepting all your input numbers, you’re off by almost a factor of 10 (and you’re suspiciously bad at rounding too).

  • Posted by Tysto

    Holy crap. In trying to disprove Musk’s numbers, you’ve actually revealed that the situation is even worse than he claims. His point is that producing a gallon of gas is a dirty process by itself in part because it uses lots of electricity. You say only 15% of that electricity comes from the dirty old grid, but that means 85% is generated in the refinery itself from petroleum products. And that’s going to be even MORE polluting than the grid mix, and unlike the grid it’s not getting any cleaner.

    If we use the exact same fuel souce mix to put electricity into EVs instead of making a gallon of gas, we could drive (nearly) the same distance WITHOUT burning the gasoline. But if we just used grid electricity, if would be cleaner still!

Post a Comment

CFR seeks to foster civil and informed discussion of foreign policy issues. Opinions expressed on CFR blogs are solely those of the author or commenter, not of CFR, which takes no institutional positions. All comments must abide by CFR's guidelines and will be moderated prior to posting.

* Required