Insane Growth By Scaling Up
Updated: Oct 6, 2020
See how solar modules got cheaper by 99%.
Follow the roadmap along with Torsten.
The April 2020 update of the International Technology Roadmap for Photovoltaics (ITRPV) will make it, again, as one of my personal highlights of 2020. See the key graph above.
The x-axis shows the cumulative PV module shipments. This is the total volume of solar modules which left the solar module factories worldwide since the beginning of time. This number is measured in the metric of “power” with the unit of “MegaWatts” (MW). 1MW equals 1,000,000 W(atts). A typical solar module with a size of around 2 square meters produces 300-400W. So, 1 MW is equivalent to roughly 3,000 solar modules.
The y-axis shows the price in US$ per Watt. The authors use the value of the US$ of 2019 so that effects from inflation do not matter in this graph.
Both axis use logarithmic instead of a linear scale. This helps to show very small and very large values in a single graph. The line going from the top left to the bottom right is the best fit for the data points. This is the learning or experience curve, a methodology that has been used for various industries for decades and studied in detail for business by the Boston Consulting Group and others.
The experience curve shows that with greater experience (for solar: cumulated power production capability by all produced solar modules) the unit cost changes (for solar: price decline for solar modules in $/W). The slope of the solar learning curve tells us: the price for a solar module decreases by 23.5% every time the total number of shipped modules doubles. The reason for this is the economy of scale and improved technology:
The economy of scale mainly stems from better purchasing power, more efficient production, less overhead, and fewer costs for distribution.
The technology of solar photovoltaic power is mainly given by the efficiency at which solar modules can convert sunlight into electricity.
The key result of this evolution? Electricity from solar now costs the same or even less than coal and nuclear power. You want examples? Here they are.
Dubai: 1.4 US$cent/kWh
Announcement in April 2020, power purchase agreement for 1.5GW solar power plant by EDF (French utility giant) and Jinko Solar (Chinese solar module manufacturer)
Germany: 4.9 EURcent/kWh
Average of all solar power tenders in October 2019
How was the economy of scale possible? In my opinion, the German legislation of 2000 can be regarded as the starting point of the modern solar industry. Guaranteed purchase prices for solar and wind-generated power gave financial security for investors to put their money into solar modules and windmill producing factories and solar and wind power plants. The economy of scale and technological development automatically followed.
There are many stories to tell around this graph. The various dents all have different stories to tell and there are different explanations for the same dent, particularly for the one that peaks in 2006.
One reason why I like this graph is because I can recapture my solar life.
Pre-Industrial Solar Age
>$100/W & 0.4MW Installed Solar Power
I came to school in 1976, solar was funky stuff for satellites and some odd applications. I had no clue that there is solar power.
$10/W & 100MW Installed Solar Power
Around 1985 I learned about solar cells from a popular science magazine. My science teacher told me that solar cells don’t have enough power, and are way too expensive. Correct, but what a vision this is, electricity from the sun! Instead, he made us build a nuclear power plant with cardboard. Then came the Chernobyl disaster and we couldn’t eat the vegetables from our garden.
$8/W & 300MW Installed Solar Power
In 1992, at university, I specialized in solar cell science and learned from a German solar pioneer that solar is great but still crazy expensive. It was good motivation for a scientist! Or was it way too audacious? Will I be able to make money with this?
In 1996 I installed solar modules in Ghana to get some hands-on experience in solar. I also installed a tiny windmill on the hills overlooking Accra. All solar and wind systems were not connected to the grid. They acted as a power supply in remote areas or as fancy back-up including batteries if the grid is down due to low water levels in the volta dam. Everybody liked solar+wind. For most, it was way too expensive, only a few could afford it. This was my motivation to start a Ph.D. in solar cells. Back in the lab.
Solar Industry 1.0
$3/W & 2,000-8,000MW Installed Solar Power
From the year 2000 onwards, due to new legislation in Germany, the solar industry developed quickly, in a spectacular way. I completed my research projects at research center Jülich, Germany, and in 2004 I joined the to-be largest solar cell manufacturer Q-Cells.
It wasn’t my salary why the prices went up for solar until 2006, even though they paid me well. In my opinion, it was that demand was way larger than supply, all triggered by favorable legislation, mainly feed-in tariffs, in Germany and some other countries like Italy and Spain.
Solar Industry 2.0
From $3/W@3,000MW down to $1/W@60,000MW
2006-2008 is the starting point of the industry as we know it today. For China, solar is now a strategic industry. Chinese manufacturers cut prices across the planet and took over the industry. Almost all focused only on scaling of mainstream technology and delivering with low barriers to customers.
From now on, supply is usually larger than demand, and more and more countries change the legislation to allow solar power in the grid. The consequence? The slope of the learning curve is even steeper: the price for a solar module decreases by 40% instead of only 23.5% every time the total number of shipped modules doubles!
The European, mainly German solar cell manufacturers, died one by one, including the company I initiated. I bet on thin-film silicon, an alternative approach to the conventional and still dominant wafer-based solar cells. We could not learn quickly enough and the built-in limit on efficiency set an end to our endeavor in 2010. But I like to think that we scared the rest of the industry for a little while which made them learn faster. Some remains of that failed technology now actually help to drive the learning curve of the conventional solar cells.
From $1/W@60,000MW down to $0.2/W@600,000MW
A single factory of a single manufacturer has now the production capacity of previously many years of global annual demand. Previously accepted efficiency limits embarrass the experts. Solar cell designs ideal for high efficiency but formerly regarded as too complicated make it into volume production. I am still part of this industry as a supplier to solar cell manufacturers. We are providing solar simulators which allow precise and cost-efficient measurement of the solar cell efficiency.
I am a co-founder and the CEO of this company called WAVELABS. In that role, I experienced how cost-centered China developed this industry in an incredible way. Every year sees many challenges, but usually, it ends with a new volume record for installed solar modules and a cost reduction not imaginable in 1985, 1992, or 2004. Glad that I embarked on that journey 28 years ago.