Shades of Green: Electric Cars’ Carbon Emissions Around the Globe

Shades of GreenThe carbon emissions of grid powered electric cars in countries with coal based generation are no different to average petrol vehicles, while in countries with low carbon electricity they are less than half those of modern hybrids.

The scale of this variation implies that the climate benefits of going electric are not evenly shared around the globe.

Downloads Available:

Full Report  |  Press Release  |  Infographic

The Carbon Emissions of Electric Cars

It is quite well understood that electric cars have the potential to reduce carbon emissions, but important to realize this potential is dependent on the type of electricity that charges the battery.  Given that the vast majority of power generation around the world is grid-tied, where a car is charged plays a large role in determining its carbon emissions.

By considering the full scope of emissions that occur in both electricity supply and vehicle manufacturing this analysis compares the carbon emissions of electric cars in twenty of the world’s leading countries.

Electric cars’ carbon emissions can be four times greater in places with coal dominated generation than in those with low carbon power.

Compares the carbon emissions of grid powered electric cars in twenty of the worlds leading countires

The legend to the right of this chart helps explain what is driving the variation between countries.  All the difference between Paraguay and India is a result of changes in the fuel mix, from low carbon hydro at the bottom to high carbon coal at the top.

In India, Australia and China coal’s dominance in the fuel mix means that grid powered electric cars produce emissions ranging from 370-258 g CO2e/km, many multiples of those using low carbon sources.

Contrast this to hydroelectric exporter Paraguay where virtually all of the 70 g CO2e/km results from vehicle manufacturing, and electric driving is significantly lower carbon than using solar power.

Comparing Electric Emissions to Petrol Cars

Electric cars’ carbon emissions can vary from similar to average petrol cars to less than half those of the best petrol hybrids.

We show this by first accounting for the difference in vehicle manufacturing emissions and then calculating the equivalent petrol vehicle emissions in terms of fuel economy, MPGUS.  Presented in this way the results are more intuitive, allowing us to compare electric car emissions with conventional vehicles in a more familiar metric.

Emissions Equivalent Petrol Car

 The legend to the right gives a rough idea of which petrol vehicle, if any, has carbon emissions comparable to an electric vehicle in each country.

In Paraguay electric cars results in carbon emissions equivalent to a 218 MPGUS (1 L/100 km) petrol vehicle.  Contrast this with India where a full electric vehicle causes emissions comparable to a 20 MPGUS (12 L/100 km) petrol car.  In China the figure is 30 MPGUS (9 L/100 km).

Based on data from 2009 the US petrol emissions equivalence is 40 MPGUS (9 L/100 km), similar to a modern petrol hybrid.  But at the speed US electricity is decarbonizing this figure is rising quickly.

In the UK, Germany, Japan and Italy the broad fuel mix of natural gas, coal, nuclear and hydro means an electric vehicle’s carbon footprint is similar to the best comparable petrol hybrid, or most efficient diesel.  In the UK this equals a petrol fuel economy of 44 MPGUS (5.4 L/100 km), while in Germany it rises to 47 MPGUS (5.0 L/100 km)

In Canada and France, where hydroelectricity and nuclear energy dominate, the petrol emission equivalences are 87 MPGUS (2.7 L/100 km) and 123 MPGUS (1.9 L/100 km) respectively.  In these countries electric cars have the potential to more than halve total vehicle emissions.

Manufacturing Emissions Matter

As vehicles become increasing low carbon manufacturing emissions matter more and more.

Our central scenario for vehicle manufacturing was 70 g CO2e/km for an electric vehicle and 40 g CO2e/km for a petrol car.  As explained in the report these estimates were made using the best available literature.

To highlight how important these assumptions are for low carbon vehicles we ran a crude sensitivity test by holding petrol manufacturing emissions constant while estimating a low (50 g CO2e/km) and high (90 g CO2e/km) scenario for electric vehicle manufacturing.

Vehicle Manufacturing Emission Scenarios

The results show that manufacturing emissions become increasingly important as electricity moves to low carbon sources.

At the bottom of the graph we can see that changing the manufacturing emissions assumptions does little to affect the equivalent emissions fuel economy.  But as we reach countries like the US, UK and Germany the difference become significant.  Finally at the top of the chart it has a very large effect.

Although we believe our central estimate of 70 g CO2e/km for electric vehicle manufacturing is a reasonable one, the more important point is to stress that in low carbon vehicles, whether hybrid, diesel, plug-in or full electric, manufacturing emissions are an important share of total emissions.

For further discussion, analysis and sources please see the report.

Shades of Green documents :

Full Report  |  Press Release  |  Infographic

  • Pingback: Eco Saver » Eco News()

  • Pingback: India named least green country for electric cars – The Guardian (blog) | Blue Planet News()

  • Pingback: Electric Cars are Very Different Shades of Green Around the World –()

  • Pingback: The Week in EVs and More (Part 2) | Earth's Energy()

  • Pingback: Le auto elettriche sono davvero a zero emissioni? | Informare per Resistere()

  • Pingback: India named least green country for electric cars | Cut Global Warming()

  • Pingback: Het schoonste land voor EV? Paraguay! |

  • Pingback: Not all EVs are created equal - Go Tech 24h - Technology News Express | Go Tech 24h - Technology News Express()

  • Pingback: On Electric Cars, Enthusiasts, and the War Between Performance and Environmentalism « Life, the Universe, and Everything()

  • Pingback: Autour de la Responsabilité Globale… » Archives du Blog » Idée reçue n°3 (suite) : la voiture électrique est forcemment écologique()

  • 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()

  • Pingback: Graph of the Day: How green is your EV? : Renew Economy()

  • Pingback: T.E.N. Episode 1: EV Etiquette, Emissions, Tesla, EV advertising, CarWings and Plug-In Day()

  • Pingback: Transport Evolved Episode 167 Show notes()

  • Pingback: Transport Evolved Episode 167: Rent-a-car()

  • Pingback: Is the Future Bright for Electric Cars in the UK?()

  • Pingback: C&F Wind Turbines Help Debunk The Myth | Clean Power Solutions Blog()

  • Pingback: ohmconnect | EVs 4ever!()

  • Pingback: Good Works | The World’s Greenest Vehicles()

  • Pingback: The Seven Greenest Vehicles on Earth | JBS News()

  • Pingback: 7 Greenest Vehicles On Earth | Pictures@News()

  • Pingback: Systems Theory and Urban Form()

  • 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) |

  • Pingback: Auto elettriche e tradizionali, quali inquinano davvero meno? | Veicoli Elettrici()

  • Pingback: » EV Emissions Reconsidered Navigant Research()

  • info

    Thanks for the report.

    Regarding Diesel:

    It requires much less refining than Petroleum so fuel production emissions should be lower. Small diesel engines can be upwards of 50% thermodynamically efficient compared to petrol’s 30% – the 1999 VW Lupo 3L and my own 2001 Audi A2 1.2 TDI (81 to 86g/km CO2 average) can achieve better than 3l/100km consumption. The latter has covered over 200,000 miles over 13 years without major overhaul – a petrol equivalent does not exist – if it did it would wear out more quickly as a diesel uses about half the RPM for the same cruise performance.

    Modern Petrol emissions are now worse than diesels, not only in CO2 but some 10 times more particulates:


    DEFRA calim UK grid mix is around 500g/kWh electricity. So what does that make a Leaf or BMW i3 CO2g/km in real world terms – I have driven the latter and managed just 80 mile range compared to my A2’s 330 miles.

    What if WVO or Bio Diesel is used?

    I still doubt that ethanol from Corn is a good idea given the low energy density cf petrol and increased corrosion issues in older petrol car fuel systems.


    • Juzer

      Does this calculation include up to 19% losses in transporting grid powered electricity? Does it include the energy losses associated with charging/discharging batteries?

      • info

        Thanks Juzer, the UK figure is for Electricity production only not transmission or ac to dc charging of batteries. In UK extra 8 to 15% losses in transmission and further 20 to 30% during charging. In winter most EVs and PHEVs have large reductions in electric range. Presently conditioning is using electricity to pre warm the batteries before travel. Again emissions not counted by EPA or NEDC

        • Radical Ignorant

          That sounds super high. So you are trying to tell me that it’s more efficient to pack gasoline on car and use generator in your pavement than to transfer current?
          EPA says it’s about 6%
          30% during charging? Pulled out of thin air. And that need to be terrible battery. And not usable one. If 30% goes as heat that mean you need to charge it almost as slow as you drive otherway it will burst into flames.

          • info

            Check out : Transmission and distribution losses – the longer the cable from the power source the greater the resistance to overcome – basic physics:'s_law
            For UK: ” significant losses in onward electricity distribution to the consumer, causing a total distribution loss of about 7.7%”

            Then at the point of charging, according to Swiss report called “BEV measurement campaign”


            (Excluding any preconditioning factors)

            Around 33% losses in charging from wall socket to battery or roughly 4/3rds input required.

            So an 18kWh battery might require 24kWh charge. BMW claim around 3.5h to
            rapid AC charge an i3 (7.2kW @240V) roughly 25kWh – which provides
            around 18.8kWh for driving but of course charge rates reduce will fill
            up so this may be inaccurate. Actual house figures from a 24h tests was
            19.3 kWh mild winter charge but battery was far from depleted at start.

          • Radical Ignorant

            But “significiant losses in onward..” – this is not from wikipedia. You are manipulating. Like: here’s the quote and here’s the link, but they have nothing in common.
            I know Ohms law since I was in high school with electronic profile.
            And other claim is bullshit: actual figures are, but battery was far from depleted, what means: I have no figures but I’m guessing.
            If you put “battery efficiency” into google you will get it’s 80-90% for li-on.

          • info

            IF google is your bible try googling :

            “significant losses in onward electricity distribution to the consumer, causing a total distribution loss of about 7.7%”

            Or to make it easier read this:


            Yes it’s from Wiki!

            Batteries hold energy they don’t make it. They are incredibly poor in terms of energy density – Why do Formula E drivers need TWO cars each? Hardly ecological motorsport! Why is the single seat Solar Impulse the size of a 120 seater Airbus A320?

            Sure they are fun projects but as with Solar Impulse last year, the batteries do wear out. Degradation increases with current demand and rapid charging. It is hugely dependant on temperature. HAve you tested one of the best BEVs – the BMW i3? I’ve tested 5 of them. All used more electricity from the grid than appeared on the Max Kappa screen. It takes more energy than the capacty of the battery to charge it fully. First 75% is fast, next 25% much much slower.

          • info

            Figures from Govt and scientific papers not simple Wiki cut and paste! Have a look on Speak EV website for links.

          • info

            Nope, independent tests in Germany and Switzerland on EV charging.

  • Pingback: Go Green On Four Wheels: Guide To Eco-Friendly Cars()

  • Pingback: Time to Buy an Electric Car? |()

  • Pingback: The Seven Greenest Vehicles on Earth – EV Central()

  • Sco Jo

    Interesting and understandable!

    • Rhydian Lewis

      this is an excellent report, and clearly makes a good effort to include all ‘upstream’ factors. The only area where I disagree is that it is taken for granted that the lifetime of an EVs is less than for ICE. In the UK the average car dies at 14 years old having covered 200K kilometers(yes many cars do more distance quicker than that but many are also small cars that never do more than 10K Km per year. We have leafs and Tesla that have easily done that distance already with no battery problems (only 5-8% range loss). Plus I believe that EVs will be far more likely to be re-used/upgraded/ repurposed. For example the Telsa batteries are made of thousands of standard cells. These cells (18650 and now 2170) can be tested individually and replaced, or recovered and the goods ones re-used. It is not a simple DIY job, but there are already companies offering this option. There will also be a very strong market in used EV drive trains, as the motors, and electronics are extremely reliable with a lifetime that well exceeds that of any ICE (possibly with some bearing replacement. I have also modeled time of day differences based on this report, which gives some very interesting results…please don’t charge your EV at peak times if you care about the planet at all.

  • Deivis Loren

    Great article. Very interesting and a lot of specific information. Thanks. I will use it for my research.
    Išmetam?j? Duj? Surinkimo ?ranga

  • matt_ecocars

    Great information source. Electric cars are on the rise, we have seen huge demand for PHEV’s as they offer the benefits of EV’s with the range of ICE’s. Until more Tesla range electric cars come available we will see PHEV’s bridging the gap. BMW are now launching a new Series 2 PHEV that looks very smart. We have seen at a new wave of used EV’s and PHEV’s now coming to the market.

    • info

      PHEVs are worse energy consumers than EVs and some ICE cars for several reasons reasons:

      Additional weight to carry around – in EV mode heavy ICE, in ICE mode heavy dead battery. Outlandish Phib get s 28 mpg (imperial gallons) real world but NEDC says 147 mpg! and that excludes pollution from dirty grid.

      Audi A3 E TRON uses more energy and petrol than A3 1.4 TSI petrol car. It also has less interior space than normal car.

  • pay for an essay

    It will surely be a good and really helpful thing for many people and for our nature as well that they can still use these kind of cars for transportation that doesn’t give bad effect to out eco-system.

  • Paulo Horta

    Ok. I have to ask.
    I can understand the maths. But someone answer me based in the real scenario:
    A coal power plant “feed” some thousands homes. This corresponds to some number of kWh.
    How much coal does the power plant burns EXTRA to compensate the electric cars being charged on the grid? Is REALLY the coal power plant producing more CO2 due to these cars being charged? Or is the power plant ALREADY burning extra kWh because it’s the way it works???

    And I’m referring to this time. If the future is electric, and the CO2 must really be reduced, the fossil fuel power plants must be replaced! So this is not a problem for the future. Scotland has already closed their last coal power plant. In the US, I read that one is being replaced by a solar panel farm!

    Anyone with the knowledge can please answer?

    • info

      Scotland has the luxury of a small population and lots of Hydro power. However, they still use fossil fuels – especially in winter. If we convert 29,000,000 cars in the UK to EVs it WILL increase grid demand hugely! We barely have enough wind/solar and natural gas and nuclear power to run the country – businesses and industry have to be paid to shut down demand at peak periods. Anything that increases ireducible nuclear waste is no better than increasing CO2 from fossil fuelled plants. NOx from such power-stations is also higher than transport or domestic sources even with minimal EVs.

      • Paulo Horta

        I believe that solar panels will partner with the increase of EVs. Cars will be charged mainly by personally owned solar “mini” power plants, that people will have in their homes. Or even at work, with parking roofs covered in solar panels.
        I fully understand that AT THIS POINT, turning ALL ICE vehicles into EVs would require lots of extra power that renewables cannot provide right now. That will have to be done gradually.

        • info

          In South of Spain that might work but here in UK very unlikely. I have the maximum permitted development of Solar PV – at best can get 22kwH in a day (a long summer day from 5am to 2200h – and my house uses a fiar bit – kettles, washing machines, cookers – even in summer there is minimal to charge a real EV with a 24kWh battery. In winter – forget it! barely 1kWh on some days!

        • Pol-Henry Bonte

          sun shines overday, overday people are away with their car to eg work so they mainly load their cars when they get home, at night … no solar then unless you have very expensive storage (cost of storage, conversion losses from loading the storage and unloading it to charge the car), so probably makes it worse

    • Rhydian Lewis

      Regarding extra coal being used Paulo. The report does an excellent job using average mix per country. I have looked at time of day variations (for UK), and have worked out the difference for charging overnight. Extra coal is unlikely to be used at night (too expensive). Whereas in the day, yes extra coal will almost certainly be thrown into the firebox for EVs, (in UK terms this means 52mpg equivalent for the day, rather than 73mpg at night). During the peak times (4pm to 7pm) it is almost a guarantee that the EV’s will be powered by mostly coal, so we’re no longer talking about an eco-option at all.
      Conclusion(UK only): my 6 year old diesel has a similar footprint to a leaf being rapid charged, and is considerably more green than anyone who charges their EV as soon as they get home from work (which is the worst possible case for the planet).

  • Radical Ignorant

    It sounds like a bullshit. Why including manufacturing and compare it to miles per gallon? That sounds as if someone was trying to prove that speed of light is variable counting it in seconds per dollars.
    Mpg is mpg – if someone is changing definitions and making things obscure what’s the point?
    “Electric vehicles can halve vehicle emmisions” – that is pure bs. How vehicle which emmits nothing can emmits half of what gasoline car emmits? They will be no longer vehicle emmisions so this statment is not true.

    • info

      Humans emit emissions not just from our bodies! Everything has an energy price. Even a cyclist is said to produce 15g/km CO2 – based on calorific intake! Even making solar panels has an energy input. The lie is you can drive a 2 tonne PHEV SUV with no environmental impact – that’s the true BS.

  • John Kechagias

    The analysis above does not seem to account the amount of energy used to produce petrol. I believe the number is about 6kWh per litre, petroleum is usually transported, distilled and distributed.
    In the case of Australia the power stations are next to the coal mines and electricity is transported along wires at a loss of 3% per 1000 km. There is an embedded energy in the wires and power stations as well, but also similarity the petrol refinery, pipe lines
    As an example the BMW i3 has an efficiency of 11 km/ KWH. It has already gone 66 km before the 1 litre of petrol is then ready to be used

    • Gyrogordini

      Are you able to cite a source for the 6kWh/litre? That would be an interesting statistic to store away…

  • Charlie Hatch

    potential misprint/error: the EV vs petrol numbers are identical (figures 1.2 & 2.3).

  • SiCorlett

    Why are EVs (70g CO2e/km) more energy intensive to manufacture than ICE vehicles (40g CO2e/km)?

    • Gyrogordini

      Good question. I fear this basic assumption corrupts all subsequent analysis and conclusions…

      • Brad Sherman

        I believe it is due to battery manufacture. This increases the amount of aluminium in the vehicle plus there are high energy requirements in the mining and processing of nickel, cobalt and lithium for the battery packs. These latter costs will presumably decrease over time as economies of scale improve and manufacturing technologies become more efficient.

        You might find the US EPA report “Application of Life cycle Assessment to nanoscale technology: Lithium-ion batteries for electric vehicles” informative. It’s not a light read but it is thorough and provides some historical perspective on how these calculations have been performed.

  • Deepak Giri

    Great explanation. But I did not find answers to my question pertaining to carbon emission generated by trucks. In India Trucks usually run on diesels. Can you please explain the ways to calculate the CO2 generated by trucks?
    Case 1, trucks carrying certain weight, say 10 MT, fuel consumption 6 KMPL and distance 1,000 KMs. Case 2, refrigerated trucks carrying similar weight, fuel consumption 5 KMPL and distance 1,000 KMs.
    How does the equation/calculation change if electric trucks are put into operation?