“Modern civilization is the product of fossil fuels. Replacing them will be the greatest collective challenge humanity has ever undertaken.” — Vaclav Smil
Here is the uncomfortable arithmetic of global energy: despite decades of investment in renewables, decades of climate conferences, and genuine technological progress in solar and wind power, fossil fuels — coal, oil, and natural gas — still supply approximately 80-83% of the world’s primary energy. This figure has barely changed since 1990.
Global CO2 emissions from energy reached record highs in 2023. The absolute increase in renewable energy generation has been real and significant, but it has been overwhelmed by the absolute increase in total energy consumption, particularly in developing countries where hundreds of millions of people are gaining access to electricity and motorized transport for the first time.
To appreciate the scale of decarbonization, consider some numbers:
Smil is not making an argument against renewables. He is insisting that we understand the actual scale of what we’re attempting.
Every major energy transition in history — from wood to coal, from coal to oil, from manufactured gas to natural gas — has taken 50 to 70 years to complete. These transitions occurred when the new energy source was demonstrably cheaper, more convenient, and more energy-dense than what it replaced. They happened in a world with a much smaller energy system than today’s.
Energy transitions are slow because the energy system is built from long-lived capital stock: power plants, pipelines, refineries, vehicles, heating systems, industrial facilities. A coal power plant built today will operate for 40 years. A gas furnace installed in a house will heat that home for 20-30 years. A steel mill designed around natural gas as a process fuel cannot simply switch to hydrogen without massive re-engineering.
The entire global energy infrastructure represents perhaps $70 trillion of capital investment. Replacing it requires replacing all of that capital — on a faster timeline than any previous energy transition, starting from a larger base, with technologies that are still maturing.
Smil is neither a renewable energy skeptic nor an uncritical enthusiast. He insists on accurate numbers. The progress in renewable energy technology and cost has been genuinely extraordinary:
This cost reduction has driven rapid deployment. In some countries — Germany, Denmark, the UK — wind and solar now supply 30-50% of electricity in favorable conditions. On sunny, windy days, some electricity grids run entirely on renewables.
But electricity is not energy — it is approximately 20% of final energy use globally (the rest is direct heat and transport fuels). And electricity from wind and solar is intermittent: it is produced when the wind blows and the sun shines, not necessarily when energy is demanded.
Storing electricity at grid scale remains an unsolved problem. Lithium batteries are superb for short-duration storage (minutes to hours) but prohibitively expensive for the multi-day or multi-week storage needed to balance seasonal variations in wind and solar output. Smil calculates that to store just one week’s worth of US electricity consumption would require roughly 200 times the current global battery manufacturing capacity.
This is not an argument that the energy transition cannot happen — it is an argument that it will require technologies and solutions that do not yet exist at commercial scale.
Given the true scale of the energy transition challenge — replacing 140,000 TWh of fossil fuel energy, including industrial heat and transport fuels, not just electricity — what would a realistic and honest decarbonization plan look like?