Running our automobiles on electricity instead of gasoline shifts energy requirements from gas pumps to the grid. What’s going to happen when significant numbers of cars are plugged in at night?
Given the increase in expected delivery of Teslas to 500,000 in 2018 and the announcements by multiple car manufacturers that they would have several fully electric cars in their lineups in the next handful of years, let’s look at a 2021, five-year view as well as a 2040 view.
What’s the net? Electric cars aren’t going to break or even dent the grid. By 2021, global electrical consumption might increase by 0.16% to 1.5%. By 2040, things start to get interesting with 5% to 45% increases in global electrical consumption, but that increased demand is still a lot less than what we are putting in with renewables annually. There is no grid in the world today which will begin to struggle with likely penetration of electric cars by 2021.
In order to play out the scenarios, we need to know how many electric cars will be on the road, how far they will be driven, and how much electricity they will consume.
- I had already built a two-scenario model with what I believed were aggressive and more aggressive penetration rates. Tesla’s latest announcement merely makes the only aggressive target more likely. I built it to determine when gasoline consumption would start to fall due to electric cars — a long time, unfortunately — but it can also be used to estimate potential demand increases for electricity. According to the model, in 2021, there will be 25 to 50 million electric cars on the road globally of the over one billion cars that will be running at the time. That’s about 2.5% to 5%. Note that there are both less and more aggressive models out there that use different assumptions.
- Average distances driven vary globally, from the highest in the USA of 21,561 kilometres annually down to the European average which is around 12,800 annually.
- Electric car gas mileage equivalent ranges from about 22 to 35 kWh / 100 kilometres.
- Finally, many electric cars will still be second cars in 2021 but many won’t. For the purposes of this assessment, let’s assume from 50% to 90% of annual mileage by owners will be done in electric cars. Obviously, by 2040, the high end is more likely.
Given these assumptions, some simple math tells us that, in 2021, global annual electrical demand from electric cars would range from a low of 35 TWh to a high of about 340 TWh, with an average of 129 TWh.
That’s a lot of electricity, but it’s worth assessing in terms of the global generation of electricity. Taking the 2012 number of 22,668 TWH, we can see that consumption of electricity from electric cars ranges from 0.16% of total annual electrical generation to 1.5%, with an average of 0.6%.
Well, that’s not going to keep any utility managers up at night worrying about how they will meet the demand in the near term.
But electric car penetration will be uneven, so some locations will see higher numbers — California, Norway, and Japan being obvious examples. Higher penetrations will mean more electrical demand in specific geographies.
Let’s take the California example and play out an extreme-case scenario. Let’s assume that 10% of all cars on the road are electric, that they are the primary vehicle of 90% of drivers, and they are all 35 kWh/ 100 km Teslas. In 2014, there were about 28.7 million cars registered in California. Let’s assume that number goes up to 30 million by 2021 for round numbers’ sake, giving about 3 million electric cars on the road. That gives an annual electrical consumption of about 20 TWh.
California currently is generating about 200 TWh of electricity annually and consuming about 260 TWh (it’s a net electricity importer). So California would see about an 8% rise in demand from electric cars in the extreme case. There is a lot of excess capacity on every grid and most of the demand will be at night, when there’s even more excess, so there won’t be any issues due to this at all. Electric cars won’t be increasing peak loads by more than a small fraction.
That increased demand is actually good news, though, as electrical demand is flat or falling outright due to efficiencies and continued shifts to post-industrial economies and rooftop solar in California is cutting into utilities’ revenues. Electrification of transportation is a rare good-news story for utilities, and one of the reasons why buying an electric car is systemically virtuous; decarbonization costs money and this gives utilities more money.
What about the Norwegian example? They already have 3% penetration of electric cars and about 40,000 electric vehicles were sold in Norway in 2015 alone. It could conceivably achieve 20% of total vehicles on the road being electric by 2021, or around 520,000 cars. Assuming European average distances driven but 100% of it in Teslas, that would require about 2.3 TWh annually, or about 2% of Norway’s annual demand of 114 TWh. Once again, most of the load is outside of peak, so this isn’t a concern. Norway is an outlier in electrical usage, however, as its massive hydroelectric capacity has led to governmental policies strongly encouraging the use of electricity in many areas where other countries use fossil fuels. Its per capita use of electricity is 3 times that of the rest of Europe.
What about Japan, a country with lots of electric cars — in fact, one of the global leaders — and a challenged electrical grid? There have been over 120,000 plug-in electric cars sold there since 2009, giving it the third-largest fleet in the world, but that’s against a backdrop of roughly 75 million cars on the road over the 127 million residents, or about 1.6% of all cars. Japanese consumption of electricity per capita is much lower than the USA or Norway, about 60% and 34%, respectively, with an annual consumption of about 989 TWh in 2012. Fukushima took a great deal of generation offline, and Japan is only now starting to regain its generation capacity. Let’s assume that Japan achieves 10% of all cars being electric by 2021, that they are used for 100% of kilometres driven, and that European average distances are driven. This would lead to a total demand for electric cars in 2012 of about 34 TWh, or about 3.4% of total generation. Even in the electricity-constrained circumstances of Japan, this level of increase of off-peak demand isn’t an issue.
Projecting out further sees greater impacts. In 2040, the model suggests a 38% to 77% electric car penetration. That would be from 868 to 1,520 million electric vehicles on the road, consuming 1,200 to 10,000 TWh annually and an average of 4,100 TWH. At that point, we would see 5% to 45% of 2012 global electrical consumption flowing to cars, if overall generation stayed flat otherwise. Median penetration of electric cars but as primary vehicles is more likely in 2040, so using the 90% of miles being electric we see about 5,300 TWH or 23%. Those are more serious percentages.
However, it’s going to be fairly easy to keep up to that increased demand, as 115 GW of renewable capacity was put into operation globally in 2015, conservatively representing about 250 TWh of annual generation. If we keep hammering in renewables at exactly the same pace — a very conservative assumption as the growth rate continues to accelerate–, we would see about 6,000 TWh of additional annual generation from renewables by 2040, and the median demand from electric cars would consume about 87% of it. If instead renewables continue to grow, it would be reasonable to see 12,000 TWH or 18,000 TWH of new annual generation. At that top end, electric cars would consume about 30% of the new generation, leaving the remaining 70% to displace coal, which it would easily do with room to spare. Even at 12,000 TWH of new renewable generation, coal could be almost completely shut down with room for the electrification of transportation, an outcome much to be desired.
Electric cars aren’t going to break the grid. Anyone who says so isn’t doing the math.
Mea culpa: an earlier version of this article slipped a decimal place on KWH / 100 KM and understated consumption.