Prices of solar photovoltaic (PV) systems follow economic laws just like other things, such as computers and cars. That does not mean that the precipitous drop in prices will slow, however.
Looking at the early history of cars, we might say that Henry Ford was a genius. In 1908, the price of a new Model T Ford was $825. But by 1914, it had fallen to $440, making the car affordable to a much larger customer base. Then, on January 5, 1914, Henry Ford astonished the world by announcing that he was increasing the minimum wages of his workers from $2.34 per day to $5.00, despite his lower prices. In today’s money, that would be the equivalent of unexpectedly giving entry-level employees raises from $11.23 straight to $24 per hour. But Ford continued to reduce the price of the Model T, and in 1916, it reached $360. The business success story just went on, and in 1926 Ford was able to put his workers on a 40-hour work week. Clearly, we might think, he had to be a man with remarkable economic ability to pull this off.
On the other hand, we might take the point of view that Henry Ford was not really a genius, but was just a little bit ahead of his competitors while market forces played out inevitably. A law of economics, Wright’s Law, suggests that the cost of cars was going to decline, with or without Henry Ford’s manufacturing innovations.
Wright’s Law works with the intuitively obvious fact that the more you do something, the better you get at it. It gets a bit more specific than that, however, and says that the cost of an object is a function of the numbers that have been made. As the number produced is increased, the cost declines in a manner that is surprisingly predictable.
Over five years ago, Plos One, an online peer-reviewed journal, published Statistical Basis for Predicting Technological Progress. It explained Wright’s Law, saying, “An early hypothesis made by Theodore Wright in 1936 is that cost decreases as a power law of cumulative production.” The Plos One paper looked at 62 products in chemical, hardware, energy, and other industries, which it found followed predictable price trajectories on the long term. They include such widely varied products as phenol, ammonia, transistors, laser diodes, PVs, and wind turbines. The paper’s projections on the prices of PVs are a bit off the mark, but the difference was within its projected error range. The underlying description of the fall in prices for products appears to be valid.
I should point out that while the drop in prices cannot go on forever (nothing does), the relationship between cost and cumulative production is often expressed in terms of the percentage of reduction in price as the number of the product produced has doubled. Clearly, as a market matures, the time to double the number that has been produced lengthens, eventually to the point that the cost reduction is not immediately obvious. Also, other influences can drive prices; in the case of cars, the prices were influenced upward by increased safety and pollution requirements.
On the other hand, some cost reductions can go on much longer than one might imagine possible. An article in Wikipedia, “FLOPS,” provides a rather stunning example of the Wright’s Law at work, as it relates to computers. A FLOPS is defined as one floating-point operation per second. The Wikipedia article tracks historically the cost of computer hardware designed for high-speed mathematics in dollars adjusted for inflation to 2017 values. The cost per FLOPS in 1961 was $152.9 billion. By 1984, that had declined to $44.17 million. In 2003, it was down to $109. Now, according to the article, it has dropped to $0.03.
One other thing to note about Wright’s Law is that it applies not only to whole systems, but to the individual components within the systems. So it applies to the installed costs of PVs, but it also applies individually to the PV cells, assembled panels, mounts, inverters, and the labor required to install them, with each on its own independent price trajectory. For this reason, the prices of overall systems will probably continue until all of the individual declines have all played out. This means that if the fall in the cost of solar panels comes up against a barrier, such as a tariff on panels, the overall cost of solar installations may continue to fall, even if at a somewhat slower rate.
Like the cost of PVs, the cost of wind power is falling. The latest of prices reported in Lazard’s Levelized Cost of Energy Analysis, Version 11 show that the cost of electricity from wind power fell below that of combined cycle natural gas gas for the first time in 2011 and has continued to decline.
But the cost of windpower has been declining at a somewhat slower pace than PVs, and this implies that the cost of electricity from solar would eventually fall below that from wind. The CleanTechnica article, “Solar Smashes Wind In First German Technology-Neutral Tender” shows that we have reached that point. It reports the results of the first German electric power auction that was open to both solar and wind power. It was remarkable because every contract auctioned was won by solar power, with wind getting none at all. Now electricity from PVs is beating wind power.
The decline in solar electricity prices is clearly driven by lower costs of the PV installations themselves, and these costs have been declining faster than anticipated. On April 11, 2018, an article at ETEnergyworld.com reported that a 5000 megawatt (MW) solar park had been approved in India.
When the US DOE announced the SunShot Initiative in 2011, it had a goal of reducing the cost of utility-scale solar installations to $1 per watt by 2020. At that time, a lot of people questioned whether this could be possible. But the prices have already gone to that level and beyond it. On April 16, the winning quote for an installation of 20 MW of solar power with an 8-MWh battery for the Andaman and Nicobar Islands came in at a price of ₹1.33 billion ($20.29 million), according to an article in Energy Storage News. That is just a tiny bit over $1 per watt. It would just about meet the SunShot Initiative’s goal two years early, with batteries included.
This brings us back to the question of where the falling prices for renewable energy and battery backup will settle down. Truthfully, we cannot tell what that prices will be or when it will be achieved. It could even fall to zero, if suppliers can derive some benefit out of providing free power. Nevertheless, we can come to some general conclusions.
First, the costs of renewable energy and batteries seem very likely to continue to fall for some time, based on Wright’s Law.
Second, the fall is predictable, as it is likely to continue as long as the market continues to grow. And that is a feedback loop, because the growth is, for the present, a result of falling prices.
Finally, since the infrastructure and commodities relating to fossil fuels and nuclear power have passed a point of achieving market maturity, the discernable downward pressure on their power is no longer clearly obvious. But their competition, which has already outclassed them, has not yet achieved that point. That being the case, we can expect that the market shares of non-renewable energy sources will continue to fall, possibly with increasing speed.
The end of falling prices for PVs is probably not yet in sight.
So my advice to those who are looking to install a few gigawatts of generating equipment is, “Forget anything that sooty, smelly, explosive, or radioactive. They are not cost effective. They are not necessary. And they are not nice.”