Do Electric Vehicles Really Emit more CO2 than Fossil Fuel Burners?

What if electric vehicles were actually worse for the environment than old-fashioned gas burners? What if, somehow, we have all been deceived by the diabolical Elon Musk and his minions in Fremont California… and hidden pollution costs associated with electric vehicles are actually making global warming worse?

This story, unfortunately, keeps cropping up in corners of the internet. How can we determine what the truth is?

To address this question requires us to consider two parts:

  • The fuel cycle. What is the “well to wheel” cost associated with operating the vehicle? Depending on the power source, that may include the cost of extraction, refinement, distribution and/or generation.
  • The vehicle cycle. What is the cost to manufacture, maintain, recycle and/or dispose of the vehicle.

The discipline that answers these types of questions generally, for all types of products (not just automobiles), is called life cycle assessment. In the case of automobiles life cycle assessment is complex. Fortunately, there is a generally accepted methodology and set of models for performing this analysis. Argonne National Labs in Illinois has been working on the Greenhouse gases, Regulated Emissions, and Energy use in Technologies Model (otherwise known as GREET) since the late 1990’s. For a given vehicle and fuel system, GREET allows you to calculate:

  • Consumption of total energy (energy in non-renewable and renewable sources), fossil fuels (petroleum, fossil natural gas, and coal together), petroleum, coal and natural gas;
  • Emissions of CO2-equivalent greenhouse gases – primarily carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O); and
  • Emissions of six criteria pollutants: volatile organic compounds (VOCs), carbon monoxide (CO), nitrogen oxide (NOx), airborne particulate matter with sizes smaller than 10 micrometre (PM10]), particulate matter with sizes smaller than 2.5 micrometre (PM2.5), and sulfur oxides (SOx).

In the latest iteration, GREET now includes more than 100 fuel production pathways, and more than 70 vehicle and fuel systems.

You could easily drive yourself mad thinking about this. To keep it simple, GREET’s output is expressed simply as emissions per distance travelled. In the United States this is expressed as grams CO2e / mile, and in the rest of the world as grams CO2e / kilometre. The vehicle cycle emissions (those produced during manufacturing) are distributed over the expected lifetime of the vehicle, and then added to the fuel cycle emissions to produce a single aggregate number.

You can see examples how this works on Tesla’s 2020 Impact Report, beginning at page 13. They compare a Tesla Model 3 charged at home using their solar and powerwall product (zero cost electricity) to grid charged. They also compare personal use scenarios with ridesharing scenarios. And finally, they compare those scenarios to an average mid-size gasoline powered vehicle.

Comparison of Tesla Model 3 emissions

The Tesla graphic shows us some interesting facts.

  • Notice that the manufacturing produced emissions for personal use vehicles seems to be dramatically higher than for vehicles used for ridesharing. The emissions are in fact identical, but because the rideshare scenario presumes a million miles travelled, when expressed in grams of CO2e/mile, the graph appears to show lower manufacturing emissions.
  • Notice also that the emissions associated with charging from the grid are dramatically higher than for solar. There is an emissions footprint for purchased electricity, that will vary depending on the utility’s generation mix, and depending on where you live. In fact, Tesla shows this in multiple graphics depicting various geographies in their report. But when you generate your own electricity from the sun, you don’t produce emissions.
  • And finally, notice that the emissions from manufacturing for the solar scenario seem higher than for the grid scenario. They are, in fact, higher. That’s because Tesla adds the emissions associated with manufacturing the solar panel and storage battery into the scenario.

You can see from Tesla’s graphic that emissions associated with using a Tesla Model 3 are dramatically lower than with the average internal combustion engine vehicle they’ve depicted. However, scenarios will vary depending on where the vehicle is manufactured, and where you live. In 2017, the 2 Degrees Institute quantified this difference for the United States and Canada. The study is old, and the underlying assumptions have improved since then, but it still illustrates this point very well. In 2017, driving just 9,000 miles in California would fully offset the embodied emissions in the electric vehicles they studied. To offset those same emissions in Michigan would take over 17,000 miles…. and 38,000 km / 23,600 miles in Alberta Canada.

So yes, electric vehicles have a higher embodied carbon footprint than internal combustion vehicles. However, the difference isn’t significant enough to warrant not switching. Whether you live in LA or Calgary, within less than 2 years that embodied CO2e difference will be erased as you power your vehicle with clean efficient electricity.