Energy efficiency is universally seen as a good thing. It is so good that it has even been formulated into an official target in the global climate community. While I agree that doing stuff more efficiently is generally a worthy and good goal, I also think that it is a limited proxy to measure our climate progress – or lack of it – and focusing on it can lead to surprising side-effects.
More With Less
Energy efficiency is often measured as the amount of economic activity (GDP) done per amount of energy used (kWh). So, USD/kWh, for example. It is a nice and simple unit, but as is often the case with nice and simple units, it can hide more than it shows.
Efficiency improvement is measured as percentage of more GDP produced per kWh of energy used, compared to the previous year. This number has traditionally been between 1 to 2 percent per year. According to IEA’s new report, the trend has turned down for the last couple years. Last year, efficiency improved by 1.2%, while in the previous year, the rate was 1.7%. IEA’s target is 3% per year, which they say is achievable.
Energy Efficiency as a Climate Mitigation Proxy
There are several reasons why energy efficiency is a slightly misleading proxy and poor target for climate change mitigation progress.
First, it doesn’t measure emissions, and emissions are what matter. As an example, moving from cooking or heating with natural gas to electricity can lower the primary energy efficiency, but since the electricity can be produced cleanly, it can also lower emissions.
Second, some climate mitigation and emissions reduction activities might hurt primary energy efficiency. If efficiency is an official target, this might discourage us from using all the means available for emissions reductions. Let’s take another example: clean synthetic fuels. If we aim to decarbonize transportation, we need to stop using crude oil-based fuels. Electrification of transportation will help, but the fact remains that we will have hundreds of millions of combustion engines in use far into the future. Those engines need fuels.
If we make these fuels from hydrogen that is first produced with electrolysis using clean electricity from wind, solar or nuclear, the end product will have perhaps half (+/- 10 %) of the energy content of the electricity used. Switching from crude oil-based liquid fuels, which have perhaps 90% of the energy content of the primary energy source, oil, that was used to make them, to something that has only 40-60% of the primary electricity that was used to make them, will hit energy efficiency heavily. If we focus on energy efficiency, this will seem like a bad thing. But it can also drop the net-emissions from transportation to practically zero – which is precisely what we should aim for.
The situation looks even more drastic if that clean hydrogen is made with nuclear electricity. First, we turn primary nuclear energy (heat) into electricity at 35% efficiency[i], and then we make hydrogen with that electricity, and then synfuels from that hydrogen. The overall conversion efficiency of primary nuclear heat to liquid fuels can be as low as 20%. Yet we could have carbon neutral fuels at potentially reasonably low cost – which is precisely what we want.
Compared to using crude oil, this would decrease our primary energy efficiency by ~300%, yet it would also decrease emissions to close to zero for the transport sector. Yes, we would use many times more primary energy in transportation as we do now, but since we could make the fuels in a carbon-neutral and environmentally-sensitive way, it would not matter that much. Hence, focusing too much on energy efficiency would discourage us from doing this. Energy efficiency is useful and should be improved where it makes sense, but perhaps it should not be a policy target by itself. The target is lowering our emissions as effectively as possible, and sometimes focusing on efficiency can get in the way of that.
[i] This percentage can increase substantially with higher temperature reactors, but 35% is a reasonable average of the current technology in use.
