Solar Learning Curve: Swanson's Law Simulator (2D)

Growth Drivers

Learning rate (cost drop per doubling) 24%

Annual capacity growth 25%

Simulation speed 3.0 yr/s

Live Readouts

Year

2010.0

Doublings since 2010

0.0

Cumulative capacity

40 GW

Module cost

$1.80 / W

Cost vs. 2010 baseline

0% lower

How it works

Solar module prices don't fall with the calendar — they fall with cumulative production experience. This is Wright's Law (in solar PV, commonly called Swanson's Law): every time worldwide cumulative installed capacity doubles, cost per watt drops by a fixed fraction, the learning rate b:

Cost(C) = Cost₀ · (C / C₀)^n
n = log₂(1 − LR)
LR ≈ 20–28% historically for silicon PV modules

Because n is negative, cost falls as cumulative capacity C rises — but only through doublings, not through time directly. Time enters only via the annual growth rate g that drives how fast C compounds: C(t) = C₀·(1+g)^t. Verified numerically: at LR=24%, one exact doubling of capacity drops cost by exactly 24.000%, confirming the exponent formula is self-consistent.

Learning rate slider — how much cheaper each doubling makes the module. Historically this stayed remarkably constant even through wars, tariffs and recessions.
Annual growth slider — how fast new capacity gets installed each year; this is what converts wall-clock time into doublings.
Solar farm grid — panel count scales with log₁₀(cumulative capacity), so the farm visibly grows fastest right after each doubling. Drag to pan, scroll/pinch to zoom.
Cost curve — a straight line in log(cost) vs. log(capacity) space is the signature of a power law; the marker shows where the current year sits on that line.

Real-world anchor: 2010 baseline of ≈40 GW cumulative capacity at ≈$1.80/W, matching the roughly 30-fold capacity growth and ~90% price collapse the industry saw by the early 2020s.

Active panel (installed capacity) Unbuilt grid slot

drag to pan · scroll to zoom

Cost curve (this learning rate) Current-year marker + trail