Cars 2: How Electric, Revisited

I said in my last post that “how electric” seems like a good early question to ask: electric car, plug-in hybrid, hybrid, or regular. The money tends to work out sort of evenly: electric cars cost more but save fuel. This means, the more you drive, the more an electric car will save you. Also, the cheaper energy you get, the more an electric car saves you. You can run these numbers yourself.

But beyond “what costs less”, there’s still some nagging guilt: “dan, the world is burning, you are buying a car, cars are very bad; you at least should get the least world-burning car.” Or, at even lesser least, I should know how much I’m burning the world.

Measuring CO2

There are more environmental harms than CO2. But I’ll start there, because it at least lets us compare.

Burning gas to go N miles produces more CO2 than plugging into a wall and getting electricity to go N miles. (thank goodness, otherwise why would we do this whole electric car thing) However, creating a giant battery produces lots of CO2 too. To compare types of cars, we should consider how much total CO2 we’re creating in construction + driving.

Battery production CO2 cost

One report in 2017 said 150-200 kg CO2 equivalent per kWh of battery. Another report said 30-200 kg1. The same energy that made the first report made another one saying that it used to be 150-200 but now it’s 61-106 (or maybe 146). Based on all this, I’m going with 100 kg CO2 per kWh.

Batteries last maybe 100k miles (meh source 1, meh source 2, but that’s how long warranties are at least). Maybe 150k?

lol, entire car production CO2 cost

We can ignore that last section, because I just found a report telling us how much CO2 it takes to build a car: 5600 kg for a gas car, 6500 for a hybrid, 6700 for a plug-in, 8800 for an electric. That feels order-of-magnitude plausible based on the last section:

Gas CO2 cost

9 kg Co2 per gallon

Car mileage

Let’s assume we’re comparing (numbers rounded):

so

Gas or Hybrid

Let’s start with a regular gas car vs a hybrid: 6500 - 5600 = 900 kg CO2 = 100 gal gas equivalent for construction. So once the Prius has saved you 100 gal, you’re ahead. Every 1000 mi, the Prius saves you 13 gallons, so after 100/13*1000 = 7692 mi, you’ve broken even. so if you’re going to drive at least 8,000mi, buy a Prius instead of a Civic.

Hybrid or Plug-in (PHEV)

Now it’s not just about miles, but electric miles. If you’re running on gas, then the Prius and the Prius Prime are the same. So the Prime (which takes 200 kg CO2 = 22 gal gas more than the Prius) is worth it once you’ve saved 22 gal gas by driving close to home. (the battery only lasts about 25 mi, so think of electric miles as “around town” miles, not “road trip”.)

Driving 1000 mi saves you 12 gal gas in a PHEV. 22/12 * 1000 = 1833. So if you’d drive at least 2,000 miles within EV range, go for the plug-in.

Plug-in or Electric (EV)

In constructing a fully-electric car, you need to use 2100kg more CO2 = 233 gal more gas. When comparing to a plug-in, the miles you save are the non-electric miles, as electric miles are the same on each.

Then, similarly, driving 1000 road-trip miles (where the PHEV would be using gas) saves 12 gal gas in an EV. To save 233 gal gas, you’d have to drive 19416 miles. So if you’d drive at least 20,000 miles outside PHEV electric range, go fully-electric. It’s a little surprising that the only miles for this decision are road-trip miles, but it makes sense: if you’re just in the city and so always within the 25mi range, it doesn’t matter if your battery has a capacity of 8 kWh or 80.

Hybrid or EV

Driving 1000 mi saves you 12 gal here again too, so now it’s just: if you’d drive at least 20,000 miles, and your only choices are a hybrid or an EV, get an EV. But depending on your driving patterns, consider a PHEV too.

for me

These numbers may seem pretty low, but I’m hoping not to drive much. I’m picturing 1000 city miles and 2000 highway miles per year, and I want to have the car for at least 10 years.

That would suggest a fully electric, but (say it with me) range anxiety, as I explained last time. Probably if we went on a long multi-charge-distance road trip, we’d rent a car, which means ignore this whole thing.

The plug-in kind of makes sense though! They’re hard to find right now, but 2000 is not a lot of city miles, even if you drive rarely, so it’d end up being worth it.

And (sadly) we’ll definitely drive more than 8000 miles, so non-hybrids are right out.

in chart form

all numbers in thousands of kg

carproduction co210k city + 20k road-trip miles co2total over 30k50k city + 50k road-trip miles co2total over 100k
honda civic5.68.914.530.035.6
toyota prius6.55.411.918.024.5
toyota prius prime PHEV6.74.311.012.619.3
chevy bolt EV8.82.110.97.216.0

in conclusion

Outside of my own decision making, this makes me think a few things about electric cars more broadly. First, there is definitely a difference in co2 impact of production, but it is outweighed by the gas you’d spend even if you’re a pretty light driver.

Second: yes, electric cars are good, plug-in hybrids are good, regular hybrids are good. But they’re not that good.

By going electric, you can cut your carbon use in half. (Maybe not even that, if the “car total footprint” report was an underestimate; maybe more than that, if you have access to super clean energy) So it’s worth picking an electric car, if you’re going to have a car! But the real way to use less carbon is to just stop driving.


  1. 2400-16000 kg for a 80 kWh Tesla battery ↩︎

  2. I’m using MPGe here to try to account for the CO2 of the electricity you use to charge. This can vary a lot too - if you have 100% free solar, then you have basically infinite MPGe; if you’re getting coal energy, it goes down a lot. ↩︎


blog 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010