Are solar power forecasts too pessimistic? Analysts missed by 59% in the past 4 years...
As Niels Bohr is reported to have said: "It's hard to make predictions, especially about the future." Yet we rely on all kinds of forecasts about all kinds of things to get an idea of what the future will be like. That's not a bad thing - it's better to have an informed guess than to have no idea at all - but we must always remember to calibrate our trust in these forecasts properly, and to go back once in a while to see how things actually turned out compared to the original projection.
A lot of environmentalists get discouraged by gloomy predictions. Oh, in 2050 clean energy will only represent a tiny fraction of the total, they say. Electric cars will stay a small niche... Deforestation will be X hectares per year... Etc. But the reality is that forecasters are only human, and they have limited data and make mistakes (sometimes big ones), especially with things that involve exponential growth, or where a single tipping point can dramatically change things. Humans just aren't very good at dealing with things that grow exponentially rather than linearly, and we can't predict turning points very well.
According to a report from the National Renewable Energy Laboratory (NREL) and the Lawrence Berkeley National Laboratory (LBL), forecasts made in 2010 about the cost of utility-scale US solar energy are off by 59%. And I don't mean that solar costs 59% more than expected, I mean it costs 59% less! In just 4 years! Imagine the margin of error over 10-20-30 years or more!
The rate of decline in costs is extremely rapid, and thus hard to predict.
between 2012 and 2013 alone, the price of a solar system your neighbor or your local businesses owner might install on their roof dropped 12 to 15 percent. And depending on where it’s located and the shape of the market, it could fall another 3 to 12 percent by the end of 2014. (source)
The larger systems built by utilities (more than 5 megawatt) now have a cost of just $1.80 per watt!
“There is still considerable uncertainty as to how low PV system prices will drop in the next five to 10 years,” said David Feldman, a lead author of the report and an analyst with the NREL. “However, there appears to be an emerging consensus that the SunShot’s price reduction targets are within reach and more and more likely to be realized. We see this reflected in the fact that many of the current projections are far lower than projections made in the recent past by the same sources.”
I've always liked the graph above, which I talk about in more detail in a post titled "This graph shows why solar power will take over the world". I think that title kinds of says it all. With prices falling this quickly, while most 'dirty' energy sources are becoming more costly, or at least not falling as fast, eventually nothing will be able to compete with solar. The only thing that will stand in the way is the intermittency issue, but that's being solved by http://www.treehugger.com/renewable-energy/grid-scale-metal-liquid-batteries-could-revolutionize-renewable-energy-use.htmlYour text to link... and smarter grids too.
That's why I take forecasts that say that electric vehicles will be X% of total vehicles sold in 2040 or whatever. How good were we at projecting this stuff 25 years ago? Sometimes single events can make big changes. GM's decision to kill the EV1 and not pursue electric cars had a big impact. That Tesla didn't go bankrupt (they came very close) in 2009 and succeeded in launching the Model S also had a big impact. If they can get to the mass-market Model 3, that'll change things too. Add or remove any single thing and it's the butterfly effect, and in 25 years you might end up in a very different place. And I'm not even talking about things like oil prices, carbon taxes, etc. That's why forecasting is so hard.