The experience curve is a useful tool for forecasting future cost reductions in solar.
This curve, also known as the learning curve, reveals a consistent pattern in solar: For every cumulative doubling of production volume, the price of PV modules has decreased at an average of around 20 percent.
It has been this way since the early days of the industry (with some ups and downs due to varying market conditions), and it continues to this day.
But there's one problem. People keep calling this phenomenon "Swanson's law," named after me.
I didn't invent it.
Over the years, my name has become inextricably linked with the experience curve in solar. This is a complete misnomer. Greentech Media has graciously offered me a pulpit to set this right.
I first heard about learning curves from Paul Maycock in 1975. He had recently joined the Department of Energy from Texas Instruments. During an interesting presentation at the Photovoltaic Specialists Conference, Maycock showed how learning curves could help guide our industry.
He explained how learning curves are a plot of the log of price versus log of cumulative production volume. (Texas Instruments had made extensive use of learning-curve projections to forecast future chip costs.)
Robert Margolis, who is currently at the DOE, covered this in detail in his doctoral dissertation at Harvard. His 2002 thesis is titled Understanding Technological Innovation in the Energy Sector: The Case of Photovoltaics. Margolis showed how almost all manufactured goods yield a straight line on a learning curve graph, with the cost typically falling about 20 percent with every doubling of cumulative volume.
The PV learning curve was helpful in making legislators more comfortable with subsidies, because it gave hope that the industry would be able to eventually wean itself from the need for government support. Sure enough, this is coming to pass today.
How did my name become linked to all of this?
During the 1990s and early 2000s I gave many presentations (as did others) that included learning-curve projections. They were meant to be inspirational -- to show solar's potential. On one occasion, I gave a talk to a group of journalists. They, not being familiar with the history, assumed that it was somehow my idea.
Whey they wrote about what I presented, they called it “Swanson’s law.” It grew virally from there.
It seems nice to have a “law” named after oneself, but in this case it has been frustratingly embarrassing. Those in the industry who follow costs know that I had nothing to do with the learning curve observation. I was just the messenger.
So that's the truth. I very much appreciate the opportunity to set the record straight.
Oh, and by the way, future module pricing will continue to decline, probably following the learning curve. The result: PV will soon become the lowest-cost source of electricity, even without subsidies.
Dick Swanson is the founder of SunPower. Listen to Swanson tell the story about building SunPower on the Watt It Takes podcast.