EVs' dirty clothes

The assumption of those buying, selling and seeking subsidies for EVs is that they are somehow cleaner than existing internal consumption engine cars.

The premise has numerous largely unexamined assumptions. One is that the alternatives are a stationery target — that there’s no progress on competing technologies (e.g. hybrids, diesels, fuel cells, etc.). Another is accounting for the energy cost of creating the batteries and the pollution cost of disposing of them.

And — as with any environmental investment — there is the question of whether this is the most efficient way to spend billions or trillions to save the planet. If (hypothetically) it takes $5 trillion to replace the global transportation infrastructure to not emit carbon but only $1 trillion to sequester carbon, couldn’t that extra $4 trillion be used to cure malaria, provide safe drinking water, or other alternatives that improve the health of the planet and its residents?

However, these are second-order arguments that don’t seem to be getting traction.

A much simpler argument is: is generating (and distributing) the electricity to the new generation of EVs cleaner than currently available gasoline-powered cars?

Increasingly, evidence suggests that the answer is “no”. As I noted last May, outside of New Zealand, few countries have green enough aggregate energy generation to reduce CO2 emissions if you shift from gasoline to grid power.

If you look a little deeper, the US picture is even worse. There are two types of electricity — cheap baseload and expensive (or more variable) peak load, used when lights are on and air conditioners are running. Solar obviously is a daytime peak load source, natural gas (the cleanest fossil fuel) is expensive and used only at peak load, and who knows when wind will be available.

However, smug homeowners with their non-polluting EVs are plugging into the grid at night, when there’s no solar. According to John Petersen (writing at Renewable Energy World) those kilowatt-hours will come from baseload power — which in the US means either coal or nuclear power. His article draws on a 2008 Rand study of the lifecycle comparison of HEVs vs. PHEVs.

The Rand study says that if you use natural gas to generate electricity, a PHEV is cleaner than a HEV but if you use coal it’s dirtier. (What about CNG cars? Petersen doesn’t say). Running coal plants to charge “clean” EVs is obviously somewhat of a contradiction in strategies.
From my analysis of 2008 EIA data, coal accounted for 48.2% of US electricity consumption and nuclear 19.6%. (In California, it’s 57.7% natural gas, 15.6% nuclear but only 1.1% coal.) Of course, shifting transportation from liquid fuels to the grid would require incremental increases in electricity generation — retiring fewer coal plants or even building more of them.

The one gap in Petersen’s analysis is that hydro can be shifted to be used whenever power is needed, so that increased electricity consumption at night could be fed by hydro. Of course, that takes away from its availability at (the more valuable) daytime peak load. More seriously, US hydro is only 6.2% of consumption and pretty much capped in absolute terms.

So this comes back to the fundamental systemic innovation problem: changing our century-old transportation system to be more green is complex and expensive. If we don’t use market forces — or distort the market by favoring one approach over another — there are likely to be suboptimal choices made.

In this case, it appears that it would be better to shift the grid to renewable energy first, and then put cars on the grid, than to add EVs to the grid at a time when renewable energy is a relatively small part (<10%) of our electricity supply. The only encouraging news is that when they spend their own money, consumers are thus far resisting the EV hype machine and going with more economically efficient alternatives.