For purely hypothetical discussion, I was asking myself how big electrical production increase would be needed for a first-world country when switching all cars from fossil fuel to EV only. I made the calculation myself for the country I live in (Italy, though the reasoning in principle would be valid for any chosen country), but the result is somehow unconvincing to me. So I would like to share my reasoning here and read you comments about it. Please note this is a very rough estimation, so please point out lack of precision in the calculation only when really matters to the general result, keeping in mind this is a very high level calculation.

In the examples I will reference data from 2019 which are the most recent complete dataset I was able to find.

The annual consumption of fuel for transportation was 7.3 million metric tons in gasoline and 23.8 million metric tons in diesel. The energy content is 46.9 MJ/kg for gasoline and 45.8 MJ/kg for diesel. The energy content sums up to about 344 PJ and 1090 PJ respectively. Converting to Wh (for easy comparison with electric consumption data) the total is just short of 400 TWh.

I found out that the efficiency of an internal combustion engine mounted on a vehicle is something around 20% - 35%, so I assumed an average value of 30% for the sake of simplicity. So the real energy that goes to motion is around 120 TWh. The efficiency of a full electric vehicle seems to me to be around 75%, so the electricity needed to charge that hypothetical EV fleet would be a little less than 160 TWh.

In the same year 2019, the electricity consumption in Italy was 320 TWh. So the conclusion of all this reasoning is that a country should roughly add 50% of its electrical production to support a full transition to electrical mobility.

Instinctively this seems way too much: have I made any serious error in my assumptions or calculations? Or is this basically a correct conclusion?

  • 1
    Is your question about switching all vehicles, including trucks? (Asking because your title only says cars.)
    – Nic
    Commented Aug 14, 2021 at 4:06
  • Thinking of country size may make a larger impact, for example when I go to see my mom it is a 600 mile trip each way if I go to see my son it is a 2800 mile trip each way. There is no public transit in my area and my daily commute is 43 miles. I have friends that have a much longer commute and some that live in the city with only a 5-10 mile daily commute so the location in the world may create a large differential on the amount if they can get battery life high enough to make commuting a reality and the power grid beefed up enough to handle all the additional charging stations.
    – Ed Beal
    Commented Aug 14, 2021 at 21:06

3 Answers 3


By order of magnitude your approximation is reasonable. I compared it to a 2021 analysis performed by Mark Jacobson as part of his 100% Wind, Water, and Solar (WWS) All-Sector Energy Roadmaps for Countries, States, Cities, and Towns. The results for Italy are available here (pdf).

For a complete electrification of the transport sector by 2050, Table 2 shows an estimated average annual demand of 19.51 GW. Over 8,760 hours per year, this translates to 171 TWh.

The difference between your estimate of 120 TWh is likely accounted for by a few factors:

  • Non-petrol loads already on the road (EVs and electric trains or buses)
  • A broader definition of transport including trains, boats, and planes
  • Growth from 2019 to 2050

In my country, a typical motorist drives 15000 km/year. With state of the art electric cars that consume less than 0.2 kWh / km, and taking into account charging losses (that bring the total consumption from the grid to about 0.2 kWh/km), that's about 3000 kWh/year for a single passenger car.

The world produces 25600 TWh/year of electricity and has about 1 billion passenger cars. So only 12% increase in global electricity production is needed to electrify all passenger cars.

However, there are other vehicles too that may benefit from electrification: vans, buses, heavy trucks, etc. So the true answer that to electrify all that can be electrified is larger than 12%. In my country (Finland), passenger cars amount to 55% of emissions from road traffic so to electrify all road traffic would require about 22% more electricity generation.

If some of the electrification is indirect through hydrogen (for example long haul heavy trucks might benefit from hydrogen because to get enough range from batteries would require ridiculous battery sizes) rather than direct through batteries, then that figure of 22% increases.

Yet, it is still manageable. Some other future uses of electricity include:

  • Heating using heat pumps
  • Production of hydrogen for steelmaking (by reducing iron ore to iron using hydrogen instead of coal) and for ammonia production (needed to produce fertilizers)

World ammonia production requires 42 million tons of hydrogen and to produce 1900 million tons of steel, 103 million tons of hydrogen are required. To make that 145 million tons of hydrogen needed for steelmaking and ammonia, 8200 TWh of electricity is needed. This is 32% increase in global electricity need.

About 50% of final energy consumption is heating, and about 50% of that heating is space heating (the rest being mainly industrial heat). World final energy consumption is 113000 TWh/a, so space heating need is 28250 TWh/a. A typical heat pump might have a COP of 3, so around 9400 TWh/a of electricity is needed for space heating purposes. This is 37% increase in electricity consumption.

So, to summarize, there are at least three reasons why electricity consumption is increasing:

  • Buildings are increasingly heated with heat pumps (37% increase)
  • Steelmaking and ammonia production require hydrogen (32% increase)
  • Road transportation will be electrified (22% increase)

Out of these three, road transportation is the least significant driver of increasing electricity consumption.

The industrial heat is tricky to estimate so I left it out. Part of it might be possible to produce by heat pumps, but part may require such high temperatures that resistance heating is the best approach. Some may require hydrogen due to even higher temperatures than resistance heating can produce.


An electric transportation infrastructure requires a 14 fold increase in grid construction.

Global demand for transportation energy (2018)

Global Energy consumption circa 2020: about 600 Exajoules (600 quintillion Joules). Transportation energy consumes 25% of global energy demand (150 Exajoules). 77% of that uses motor gasoline and Diesel fuels (Cars and trucks mostly) or 115.5 Exajoules (subtracting some because gas and diesel are also used in agriculture, construction, lawn equipment) that's....probably averaging 80-90 Exajoules.

Conversion to units of electricity (watt-hours)

90 exajoules/year = 25000 TWh (or 68.5 TWh/day). Averaged over 24 hours, that's 2.85 terawatts of constant power generation. Factor in transmission losses and thermodynamic energy losses in generation and consumer grid and waste heat, probably 3+ million megawatts. That's 3000 Nuclear reactors, just to replace the automotive/truck fleets energy demands. Also to keep the grid from melting from all that extra power, you'd have to build the grid out 14 fold to handle all that juice.

  • I downvoted because this answer ignores the fact that electric vehicles (as compared to internal combustion engine vehicles) require substantially less primary energy per distance travelled.
    – Nic
    Commented Aug 16, 2021 at 21:29
  • In the use stage maybe! Cuz you forget charging an EV quickly uses more power. A 100 kwh battery half depleted needs 50 kwh to charge. In 8 hours needs 6.25 kilowatts, To charge in 30 minutes as they promise needs 100 kilowatts! This massive power demand places huge demands on grid to have stability to prevent a blowout, and cooling, etc. More power is needed in distribution. From generator, transformer to end consumer 1/3 of electricity will be lost as waste heat. And EV efficiencies depend on temp, humidity. Some vehicles lose 20% of battery in casual/traffic driving
    – LazyReader
    Commented Aug 17, 2021 at 8:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.