The ‘electric cars aren’t green’ myth debunked

by Lindsay Wilson in Travel

Electric cars green myth

It’s time to bust this thing wide open.

‘Electric cars aren’t green’ is a great bit of counter-intuitive headline bait, but it’s bad maths.  This is how the argument goes, again and again…

Electric cars have higher manufacturing emissions than normal cars.  Electric cars also use electricity that has its own footprint.  And put together these two factors are a ‘dirty little secret‘ that negate any climate benefit of electric cars!

No.  Let’s clear this thing up once and for all.

It’s all about the juice

One of the most irritating things about articles discussing electric car emissions is the way it’s always very black and white.  In one corner you have the ‘zero emissions’ brigade and in the other the ‘worse than combustion engine’ crew.

But as ever, real life comes in shades of grey.

The reality is that even after you account for the bigger manufacturing footprint of an electric car it is all about the fuel mix of the power you use, the ‘juice’ if you will.

Using coal powered electricity electric cars do nothing to cut emissions, using natural gas electricity they’re like a top hybrid and using low carbon power they result in less than half the total emissions of the best combustion vehicle, manufacturing included.

In our recent study ‘Shades of Green: Electric Cars’ Carbon Emissions Around the Globe‘ we calculated grid powered electric car emissions in twenty countries. But we actually had data for quite a few more countries we didn’t include.

So let’s break out the data and put this thing to rest.

Mapping electric car emissions

The following map compares the carbon footprint of electric driving using average grid electricity in 40 or so countries.  The actual carbon intensity of electricity you use may differ from the national average for a number of reasons, but it’s a great starting point.

The results are shown in terms of grams of equivalent carbon dioxide per vehicle kilometer (g CO2e/km).  Each estimate includes emissions from vehicle manufacturing, power station combustion, upstream fuel production and grid losses.

The specs are based on a full electric vehicle, similar to a Nissan Leaf, using the 2009 average fuel mix in each country.  For each country vehicle manufacturing emissions are assumed to be 70g CO2e/km, based on a number of studies detailed in the report.

EV emissions by countryClick image to expand

Of the 40 countries covered in this map emissions vary from 70g CO2e/km in hydro loving Paraguay, up to a 370g CO2e/km in heavy coal using India.  The US average is 202g CO2e/km, in China it’s 258g and in Canada 115g.

In Paraguay virtually all the emissions are from vehicle manufacturing, as the power is incredibly low carbon.  Whereas in India the breakdown is 70g for vehicle manufacturing, 200g from power plants, 30g for fuel production and a whopping 70g for grid losses.

The colors in the legend split the countries into five different groups based on carbon intensity.  As you can see, even after vehicle manufacturing is included the carbon intensity of driving an electric car varies 5 fold based on the juice.

For a bit of reference, the average American gasoline vehicle is up at about 300g CO2e/km, while a new hybrid might manage 180g CO2e/km after you include vehicle manufacturing, fuel combustion and fuel production.

Compared to combustion vehicles

Because grams per kilometer is such a funny metric it is nice to convert these results to something more familiar.  Working backwards from the data we can estimate what type of conventional vehicle (if any) would produce similar emissions.

For want of a better phrase lets call this the ‘Emissions equivalent petrol car’.

EV emissions equivalent

Click image to expand.

Now the figures are much easier to get a grip on.

In coal heavy India, China, Australia and South Africa electric cars using grid power are just like typical gasoline vehicles, in the 25-30 MPGUS range.  In the UK, Germany, Japan and Italy they are as good as the best petrol hybrids, in the 45-50 MPGUS range. But in low carbon supply places like France, Brazil, Switzerland and Norway they are in a different league, averaging well beyond 100 MPGUS for equivalent emissions.

It is important to remember that the electricity you get might not match your national average for any number of reasons.  The night time intensity might vary, you might have solar panels or live in a country like the US, where the grid is actually a bunch of separate grids.  For example in Colorado a grid powered electric car is equivalent to about 30 MPGUS, whereas in California it’s up around 70 MPGUS.

For all the comparisons in this map the vehicle manufacturing of a gasoline car is just 40g CO2e/km compared to 70g CO2e/km for the electric vehicle.  This is because we have accounted for both a greater manufacturing footprint and lower lifetime mileage in an electric car.

If you are interested in the detail check out the full report.  It includes a breakdown of all figures, sensitivities to manufacturing, vehicle performance and comparisons to diesel vehicles.

Electric cars are as green as their juice

Critics of electric cars love to talk about manufacturing emissions and putting horses before carts.  But they never seem to offer any better solutions.  If they were waxing lyrical about urban densification, electrified public transport and the joys of bicycles their critiques would ring true, but that’s not what you hear.

Electric cars are relatively new at a commercial scale and are dealing with issues of cost, range and charging speed.  Each of which will be helped by improving batteries.  Despite this they offer enormous hope for reducing carbon emissions, improving local air quality and limiting noise pollution.

Electric cars are far from perfect, and there are plenty of valid ways to critique them.  But let’s not pretend that a gasoline vehicle can compete with an electric car in terms of carbon emissions.  It’s just not a contest.

Give an electric car the right juice and it crushes combustion engines.


Download the report here: Shades of Green 

Author’s note: in case you are wondering, I don’t own an electric car. We have an efficient Skoda diesel which is mostly used at weekends with 4 people in it.  My preferred mode of transport is my old dutch bike, which in terms of gCO2e/km trashes all comers (foodprint included).

Don't forget to grab your free copy of our eBook, Emit This.

  • Parax

    Emissions are not just CO2, It would be good to include CO, O3, NOx, THC, PM10 in Emissions discussions also and contrast Street Level Emissions verses Grid Emission and its impact on Public Health.

    • Lindsay Wilson

      So true. To much for me to get into hear I’m afraid. Also particularly important when assessing the benefits of diesel cars in Europe. Generally an argument for electric cars, particularly in cities. Although with the wrong power source this again creates problems:

  • Pingback: Infographic: The Real Emissions of Your Electric Car Will Depend on Where You Live | Inhabitat - Sustainable Design Innovation, Eco Architecture, Green Building

  • Pingback: "Electric Cars Aren't Greener" Myth Debunked

  • Craig Wolfe

    Lindsay – All of your data and and all of your writing are very helpful. I will keep pushing you out there. Concert for the Climate Guy.

  • Pingback: Graph of the Day | Earth's Energy

  • Pingback: 5 Elements of Sustainable Transport | Asia Pacific Sustainable Transport Foundation

  • Pingback: Graph of the Day: Five elements of sustainable transport : Renew Economy

  • Pingback: Energy production methods dictate where electric cars are the cleaner option (or not)

  • Pingback: 5 Elements of Sustainable Transportation | Enerji Medyas?

  • Pingback: What Do Paraguay, Norway, Iceland, And Brazil Have In Common? This Map Will Show You. | Omaha Free Press

  • Pingback: What Do Paraguay, Norway, Iceland, And Brazil Have In Common? This Map Will Show You.

  • Pingback: 5 Elements of Sustainable Transport | Pictures@News

  • Mike Speck

    Great info Lindsay. Appreciate the clear concise comparisons. One thing though. Do you have any data on the effect on the environment of the spent Lithium ion batteries once the cars are “put out to pasture?”

    • Lindsay Wilson

      Sorry, had missed this. I’m not really to sure about this stuff. I know that the batteries can be used in lower grade situations when they are no good for cars anymore, but not too knowledgeable on the recycling front

    • Randy

      In Japan or Australia (maybe both, i cant really remember) a bunch of old EV batteries were recently put into service as grid storage. They will operate for many more years until they are too old to be effective anymore. They will then be recycled, with over 90% of the materials being reused (generally stuff like the plastic casings in the batteries is not recycled because it isnt cost effective from the company’s standpoint). Tesla Motors plans to integrate battery recycling into their battery manufacturing plant, where the batteries will be made into new batteries.

      Currently 96% of a lead acid battery is recycled. A battery that weighs 10x or more and is more valuable (still worth over a thousand dollars at the end of life in the vehicle) will surely be recycled, especially since the batteries can not be removed easily…

      The batteries should last a long time, as long as a car today is expected to last. In 10 years, a battery replacement for a Nissan Leaf should only cost 2 or 3 thousand dollars and the car should keep on going. Unlike in a conventional ICEV, not much can go wrong in an EV. So EVs should last much longer.

  • Pingback: The Black and White Facts of "Green" Electric Cars - Ecodriving Solutions

  • Pingback: The Electric Car Debate | Clean Fairfax Council

  • Pingback: Systems Theory and Urban Form

  • Esther C

    I understand that the source of our electricity determines the emissions from driving EVs.

    But I think that in order to approximate the environmental impact from manufacturing, we need to know: where/how is the lithium (and other minerals and metals) mined, is it shipped elsewhere for battery assembly, and then is the battery shipped again somewhere for vehicle assembly?

    Consumers need to be able to estimate the overall impact. Just because I live in Vancouver, and ultimately driving an EV is very clean (we are essentially 100% hydro electric) — I can’t use your “made in Canada” electricity stats to calculate environmental impact from manufacturing.

    Is there a list, anywhere, of EV cars and where the batteries are made and cars assembled – that might provide a somewhat accurate estimate?


    • Lindsay Wilson

      Links to the three studies I used to reference the manufacturing emissions are in the report. At the end of the day vehicles have a global supply chain. Lithium comes from Chile, Australia, China . . but the primary driver of emissions is process not transport. Here’s the study: I’m not aware of a list of assembly locations, though it may exist

    • Randy

      Nissan leaf battery and car is made in Tennessee (some battery materials come from Japan, a lot of Leaf materials come from Japan)

      Tesla Model S battery cells are made in Japan, packaged in California, and majority of the car’s components are North American (Canada or USA) and the car is made in California

      Volt battery is made in Michigan. Car is built in MI. I think some cell material comes from Korea, not sure.

      I think that the new Smart EV is different, but I believe last one was made in France and the battery cells made in Japan, assembled in France, probably…

      Most lithium comes from South America. There is only about 9 pounds of lithium in the Leaf. As far as I know, there is only one lithium production operation in the United States. I dont think there is any production/mining in Canada, not sure…

      A gasoline vehicle has a lot of heavy medals, expensive metals (like platinum), and rear earth elements. These often come from China. The Model S and Chevy Spark EV do not use rear earth elements (although the touch screens may have some in it, if it is capacitive, then likely does)

      The batteries are fully recyclable and will likely be used in NA for grid storage after the battery is replaced. Then the batteries will be recycled into new batteries.

      EVs dont require transmissions, emissions equipment, ICEs, oil changes, filters, and many other bits. That is a LOT of savings. Lithium does not make up much of the battery. Over the life of the battery, it would take many times more (think over 100x) motor oil to run a conventional ICEV than it would take lithium to make that EV battery… And that battery will be reused then recycled.

  • Ricardo Fraguas

    Thank you very much, Lindsay. It is very clear. More than ever, let’s generate clean energy at home, let’s push the eradication of fossil fuel use for energy production and promote renewables. when driving, let’s drive electric.

    Best R.

  • gasdive

    You’ve included the CO2 generated in the manufacture of the electricity when you compare petrol cars with electric. Why then have you not included the electricity required for the extraction, transport, refining, transport and dispensing of petrol for petrol cars? Given the revolting stuff that crude oil is and the lovely clear pure liquid that petrol is, it’s got to have a fair bit of processing. Paper which is just wood ground up with water and then rolled flat uses 7 kWh/kg. The lowest plastic is twice that. So conservatively 7 kWh/litre (a bit less than a kg so slightly more than paper and much less than plastic). The Leaf has a 24 kWh battery that takes it 160 km. My petrol car uses 16 litres of fuel to cover the same distance. That’s 112 kWh of electricity to refine (ignoring extraction and distribution) the petrol to cover the same distance that the leaf covers on 24 kWh. To say as you do: “Using coal powered electricity electric cars do nothing to cut emissions” is simply, and obviously wrong. If you include the extraction (from some miles down) and the various pumping that occurs into and out of tanks or through many km of pipeline, (much of it as very viscous crude) then you’re even *more* wrong.

    • Lindsay Wilson

      I do the full life cycle for both. It’s all shown in more detail in the study

    • Randy

      It depends on the oil, but I read that the best “conventional oil” has about 7kWh of electricity in it per gallon of gas, while the worst case scenario, tar sands oil, is about 22kWh per gallon of gasoline. But its hard to find good sources for the amount of electricity it takes to refine and transport oil and its fuels…

      • Lindsay Wilson

        The figures in the report are made using conventional oil. If you add in the amount of tar sands oil the equivalent MPG figure jumps by 3 in the US

    • Lindsay Wilson

      I have included the full lifecycle for both the gasoline and electricity pathways. Please see the report

      • gasdive

        I read the full report. You use the figure of 0.46 kg of CO2 per litre of fuel production and distribution. You keep that as a fixed level for all fuel production regardless of where it’s refined. I’d suggest that 0.46 kg of CO2 per litre is far too low. Your figures are that in India 300 g of CO2 is emitted to make 211Wh of wall electricity. So you’re saying that to make a litre of petrol in India requires less than 324 Wh electricity. I say “less than” as there are other carbon emissions related to fuel refining that are not electrical in nature. Can you point to *anything* that would indicate the amount of electricity required to refine and distribute fuel is so low? It’s hard to say for sure as the oil companies are very coy about revealing the amount of electricity it takes to refine fuel. However figures are available for other products. Paper takes 7kWh/kg which is about 20 times more than your figures for Indian fuel production. Plastics like PVC (more similar to petrol than paper) has about 21 kWh/kg or about 65 times more than your estimate. I’d have to guess that fuels come somewhere between paper (ground up wood soaked in water and rolled flat) and PVC. I’ve worked in an oil refinery and while I wasn’t privy to the amount of electricity it used, the power lines that came into the plant were huge high voltage transmission lines bigger than the ones that serve large towns. I feel the figures you’ve used (which admittedly were not yours) are absurd. The Fawley refinery in the UK has it’s own power plant with a capacity of 135 MW. They don’t use it to toast chip butties.

        • Lindsay Wilson

          I’m using figures based on meta studies of numerous lifecycle analyses across different countries. Yes, these figures will vary across refineries and countries, but the variation isn’t too large. I can’t see the point in using kWh metrics when a whole literature of lifecycle analysis is available. So many of the energy inputs are non-electric

          • gasdive

            I’m not following how you can have a whole essay on how the amount of carbon emitted by an electric car varies depending on which country it’s in but you then say of carbon emitted due to the electric inputs into oil refining (at least 50% as large as the electric inputs into electric cars but more likely more than 100%) the “variation isn’t too large”. I agree with you that “many of the energy inputs [to oil refining] are non-electric” but just the electric ones *alone* are at least a sizeable fraction of the electricity used for electric cars. I’m not sure how to phrase it differently and I’m just repeating myself now. “Petrol cars use at least 50% as much but quite likely much more electricity than electric cars”. If you’re going to count the CO2 used in the production of electricity for electric cars, you should count the CO2 used in the production of electricity to make petrol. Your figures appear to show the CO2 released directly during the refinement of oil but neglect the CO2 produced to make the electricity used to refine the oil. Randy’s figures below of 7 kWh/gallon (we can assume US gallon given him calling it “gas”) equates to around 2 kWh/litre. For a country using coal that’s about 2 kg of CO2/litre of fuel, 4-5 times your estimate of 0.46kg/l.

            Looked at another way, the 2 kWh needed to make a litre of petrol will run an electric car about 10 km. That’s about the same as the distance a litre of petrol will drive a car. So no matter how much carbon is needed to make the electricity, the petrol car will be at least the tailpipe emissions worse than the electric. That’s even before you add in the 0.46 kg of non-electric inputs.

            I agree whole heartedly with your website, what you’re doing, how you’re going about it. The only detail I disagree with is that you’re far too soft on the emissions from refining petrol for petrol cars.

          • Lindsay Wilson

            I take your point about the variation that would result from differing locations of refining, that’s fine. My point is simply that when I looked at a whole number of upstream lifecycle analysis for gasonline from DEFRA, LBNL, EPA . . . there wasn’t much variation in the figures. My understanding for that is that many of the inputs are for extraction, or not electric, and that the refining variation isn’t too great. From what I could see much more variation resulted from whether it was conventional or tar sands oil. When I see multiple lifecycle analysis from different independent government bodies around the world in close alignment I’m not going to spend my time deconstructing them

          • Lindsay Wilson

            I looked a little more, it seems that majority of refining energy inputs are non-electric by products of refining, eg gas and coke. That may explain why I found limited variation

          • gasdive

            I’m going to split some hairs, please don’t take it as a critique. I just want to nail down the figures.

            I now get what you mean. Given that most of the CO2 generated from refining a constant, coming as it does from the oil that’s being refined, the differences between locations should not be *as* important as they are for electric cars.

            I’ve read through the link you provided but I can’t find a point where they talk about imported electricity. I can see at the end where they say gasoline has 20g of CO2 emission per MJ well to pump (fig 4) which equates to about 0.6-0.7kgCO2/litre of petrol.

            During the introduction it says that 80% of the energy required for refining comes from the feed stock. The 20% remaining is 4-7 times more than the figures that you’ve found in the paper. It also says that oil refining consumes 7% of the total energy budget of the USA. So the electricity consumption should be 20% of 7% of the total US energy consumption. Yes?

            Current US energy consumption is 25 155 TWh (2009). 25155×0.2×0.07 is 352 TWh or 3.5×10^14 Wh. The USA consumes about 5×10^11 litres of fuel per year. 700 Wh per litre.

            So for fuel refined in India or Australia that’s 995 g of CO2 per litre in *addition* to the 700 g per litre that comes from using feedstock to drive the refining. Wow, the CO2 released by making the fuel is twice the weight of the fuel!

            When it comes down to it I’m nit picking.

          • Lindsay Wilson

            There are some bizarre assumptions in there. I’ve seen refining emissions figures from 5-13 g CO2/MJ. The variation is not coming from electricity, it is coming from flaring practices and efficiency. For example Fawley refining emissions are 10% lower than Coryton largely due to the CHP at Fawley. Flaring practices (or lack of) are a big deal

      • guest

        I agree with your approach and that is how it should be looked at. pls share the report.

      • gasdive

        Lets look at it another way. The Fawley Oil Refinery has it’s own power plant that makes 135 MW. That’s 1.1×10^12 Wh per year. Divide by 211 gives you the number of electric car km that it could charge. It’s 5.6×10^9 km. The average car in England drives 8200 miles or 13000 km. That means that the power station that refines oil could if used to power electric cars instead cover the needs for about 400 000 electric cars.

        It produces a million gallons of motor spirit per day. At 20 mpg that’s enough to cover the needs of just under 900 000 petrol cars.

        So looked at that way, if you assign any CO2 to electric cars you really should assign at *least* 50% of that figure to petrol cars on top of their tailpipe emissions. If you do that you can’t say that in the worst countries the CO2 emission of an electric is similar to a standard car because the electricity used to refine the fuel for the standard car has itself emitted far more CO2.

  • Pingback: The EV Wedge: How Electric Vehicle Fuel Savings Vary By Country (and Car) |

  • Pingback: The EV Wedge: How Electric Vehicle Fuel Savings Vary By Country (and Car) |

  • guest

    It all depends on what is the source of electricity and what is the boundary/scope.
    If we talk about the total life cycle impact of an electric car it will not be as green as we think. Of course if the electricity is generated using renewable energy sources then the impact will be much less.
    The answer lies in the tapping the solar energy which is the ultimate source of energy in the universe.

    • Randy

      Geothermal and nuclear power both use non-solar energy. And the sun is only good at giving our solar system energy, not the whole universe… But EVs are the best vehicles on the road.

  • Randy

    Using coal powered electricity electric cars do nothing to cut emissions, using natural gas electricity they’re like a top hybrid and using low carbon power they result in less than half the total emissions of the best combustion vehicle, manufacturing included.

    STRONGLY disagree. They DO cut emissions. An EV in the US on 100% coal still puts out less CO2 per mile than the average car, for one. Two, you are ONLY looking at MPG. If a large city goes electric, then the smog problem likely disappears with the ICE. No more oil spills. No more gasoline is spilled at the gas station every day. There are a LOT of benefits to EVs. The grid is getting cleaner, oil is getting dirtier as we head toward oils like tar sands. And countries with high levels of coal usage have low emissions standards on cars. This means an EV will save a lot of smog and other non-CO2 emissions.

    This is because we have accounted for both a greater manufacturing footprint and lower lifetime mileage in an electric car.

    That makes no sense. EVs are virtually zero maintenance and can last MUCH longer than an ICEV. The battery may be replaced in over a decade, which should increase range significantly, last longer, and the car will keep on going. The batteries are fully recyclable and can be reused for grid storage. EVs should last LONGER than ICEV.

    • Lindsay Wilson

      Please click through to the report. The production of EV batteries is hugely carbon intensive using current practices. This may improve in the future, but its just a fact, verified by endless studies

      • Randy

        The “and lower lifetime mileage in an electric car.” was what i was strongly disagreeing with.

        • Lindsay Wilson

          This is a standard assumption for these studies. If you want to assume longer lifetime milege you then have to account for the manufacturing footprint of a replacement battery, or a much bigger battery (tesla)

          • Randy

            The cars will not be scrapped when the battery is replaced. That battery will be reused then recycled too. EVs should last longer because they do not have all the complex internal combustion engine bits that start going bad.

          • Lindsay Wilson

            Sure. But the new battery has a footprint of about 5t CO2. Hopefully this will get better with the likes of tesla going after their manufacture

          • Randy

            And that battery will have a full life, either as a source of energy for the car, or as storage for something else.

  • Pingback: Electric Vehicle Wedge -- Electric Vehicles Vs Gasoline Cars By Country & Car | CleanTechnica

  • Pingback: Electric Vehicle Wedge — Electric Vehicles Vs Gasoline Cars By Country & Car | NHAT NET

  • lauralouise90

    I’m so glad you’ve wrote this – I’m pretty on the fence when it comes to electric cars or standard battery cars but it does frustrate me how both parties are very black and white in why their way is better…. nothing is ever black or white, the very nature in how people drive is different so emissions for example will always vary from person to person!

  • Pingback: Placa Verde | Rey Peatón

  • JakoAnoniman

    While this CO2 newspeak is all so amusing, could someone please explain why a new battery for a Nissan Leaf costs me (and the unsuspecting tax payers) US$100 per month of a battery replacement program, which is equivalent to 700 miles or so in a normal car, charging not included. It runs for only ~50,000 miles between changes, if you are gentle on the pedal.
    The paper cited in your report says that “Li-ion battery plays a minor role in the assessment of the environmental burden of E-Mobility”, but at battery cost this high … not so much. It is comparable to the fuel cost for the same mileage, yet without recharging.
    CO2 or not, money-wise your calculation sucks. Big time.

Previous post:

Next post: