Energy source cost, risk and pollution

I collected numbers on the deaths per TWh of energy production for the major energy sources from Markandya (2007) (mainly fossil sources) and Sovacool (2016) (mainly renwables); numbers for the typical cost per MWh from here, and numbers for the emissions in grams of CO₂-equivalent emisisons per kWh from UNECE (2020). The results are plotted below.

CO₂ emissions vs. cost

This compares how expensive vs. life-cycle pollution (in CO₂-equivalents) the energy sources are. Note the logarithmic x axis! There are two clear groups: nuclear, hydropower, wind and solar cost roughly $60/MWh and emit about 10 grams of CO₂-equivalents per kWh (though solar is a bit higher than that), while gas and coal cost roughly 50% more and have roughly 100x as high lifecycle emissions.

Deaths vs. cost

This compares how expensive vs. deadly the energy sources are. Note again the logarithmic x axis! There are again two clear groups: nuclear, hydropower, wind and solar at around $60/MWh with roughly 0.02 deaths per TWh; and gas and coal at roughly 50% higher cost and >100x as many deaths (though coal has around 1000x as many!).

These numbers include deaths from pollution, though the nuclear numbers do not include Chernobyl since that type of reactor is no longer in use, and the hydropower numbers don't include the Banqiao dam. It's hard to estimate the deaths from Chernobyl, but Markandya (2007) gives 0.074 deaths per TWh when including Chernobyl, which is still comparable to the renewables.

Data file

Here's a data file with the numbers.