the curve that matters

how a four line model shows the real energy transition

18-Jul-25

People argue about the energy transition like it is a theology fight. Numbers help. A four line model helps more.

Start with growth. The world economy has added about 3% per year. Efficiency has clawed back about 1% per year. Net energy demand has grown around 2%. Now place clean energy on a separate track. It starts at roughly one third of final energy, then grows several points faster than demand. That is it. No mysticism. Two curves, one overtaking the other.

In the early years, this feels like failure. Energy demand rises, so even with fast growth, clean energy cannot catch all the new demand. Fossil use still grows in absolute terms for a while. Critics point at the fossil share and say nothing changes. They are looking at the wrong part of the curve.

Keep running the model. The moment clean growth exceeds net demand growth for long enough, the crossover happens. First the fossil share falls below half. Then the absolute amount of fossil energy peaks and begins to decline. Later it falls fast. Compounding does the work. We see the same effect in startups, in user adoption, in cost curves. The beginning looks flat. The end looks obvious.

The debate about primary energy distracts people. Fossil heat gets counted on the input side, while electrons and heat pumps get counted on the output side. That makes fossils look bigger than the services they actually provide. Use useful energy, not fuel input, and the picture changes. Clean starts higher, and the crossover arrives sooner.

The lesson for builders is simple. Aim for growth in clean supply that beats demand growth by a few points, then hold it. Projects that remove friction matter more than projects that demand perfection. Transmission that gets built. Siting that earns consent. Manufacturing that scales. Software that smooths peaks. If these move, the curve bends.

The lesson for policy is similar. People punish plans that raise costs, especially for those with the least buffer. People reward plans that lower bills and raise reliability. Support the things that compound, like learning-by-doing, and protect the things that unlock siting and grids. You do not need every argument to win. You need the compounding to continue.

If you accept this logic, the fight shifts. It is not about proving that fossils vanish tomorrow. It is about making the crossover inexorable. That is a better target. It survives noise, it survives politics, and it gives builders a clear task. Keep clean growth above demand growth, and keep it there long enough. The rest follows.