Learning Curve Concept

Solar panel costs have fallen 99% since 1976. This was not an accident and not primarily a breakthrough. It was a learning curve: every doubling of cumulative production reduced costs by roughly 20-25%. The learning curve is the most powerful force in energy economics, and misunderstanding it leads to systematically wrong forecasts.

A learning curve (also called an experience curve) describes the empirical relationship between cumulative production volume and unit cost. The "learning rate" is the percentage cost reduction per doubling of cumulative output. Solar's learning rate has been remarkably stable at 20-25% per doubling for over four decades. Lithium-ion batteries show a similar pattern at roughly 18% per doubling. Wind turbines have a lower rate of about 15%.

Forecasting with the curve. In 2010, cumulative solar installations were about 40 GW and module cost was roughly $2/watt. By 2023, cumulative installations reached 1,600 GW (roughly 5 doublings). At a 22% learning rate: $2 x (0.78)^5 = $0.58/watt. Actual 2023 cost: roughly $0.25/watt, beating even the learning curve prediction due to manufacturing scale in China.

Why did the IEA and EIA consistently underestimate solar growth?

Because their models assumed static costs. Traditional energy forecasting projected solar costs as flat or slowly declining. Learning curves predict that deployment itself drives cost reduction, creating a feedback loop: lower costs drive more deployment, which drives costs lower still. Every major energy agency underestimated solar for 15 consecutive years because they modeled costs without this feedback mechanism.

The curve does have a floor. Manufacturing costs compound toward zero, but silicon, glass, and lithium do not: at some point raw materials set the price, and solar modules are already approaching that point.

---